Extreme is, at its core, a cloud networking infrastructure company. It builds and sells the switches, wireless access points and campus networking fabric that underpin enterprise networks around the world. Increasingly, however, Extreme is shifting the operational experience of its customers from traditional network management software towards a cloud-native platform called Extreme Platform ONE. Delivered as a subscription service, it unifies network management, AI, automation, security, licensing and multi-vendor (third-party) management into a single operating layer.

On the surface, this year's Platform ONE announcements read like a modern AI platform release. Agentic AI, conversational interfaces, an AI marketplace and new consumption models all feature. Several of these, particularly the built-in consumption guardrails and commercial controls, appear genuinely well considered. But I don't think the long-term significance of Platform ONE lies in any individual AI feature.

The more interesting question for me is whether Extreme is becoming a SaaS company, or a hardware company with SaaS economics. The latter is far more interesting because it changes the questions buyers should be asking. Rather than simply comparing hardware specifications or AI feature lists, organisations should be asking which platform will deliver the greatest operational value over the life of the network.

In that context, Platform ONE shouldn't simply be judged on AI capabilities or subscription growth. It should be judged on whether it increases the lifetime economic value of every switch, access point and fabric deployment. If the software can continually reduce operating costs, automate management and improve customer outcomes over time, then the value of the hardware extends well beyond the day it is installed. It becomes the foundation on which an ongoing software relationship is built.

Share

Extreme still derives most of its revenue from selling networking infrastructure such as switches, wireless access points and campus fabric. These are mission-critical products that must be designed, manufactured, shipped, installed and supported over many years. That matters because enterprise networking is not a three-year software subscription. Buyers are making infrastructure decisions that often last the better part of a decade, and they need confidence not only in the technology, but in the vendor’s long-term commercial and supply chain viability.

That also means Extreme cannot, and arguably should not, be evaluated as though it were a pure software company. Enterprise networking carries a fundamentally different economic model to enterprise software. Vendors must invest simultaneously in silicon roadmaps, hardware engineering, manufacturing, logistics, global support, channel ecosystems and, increasingly, cloud software platforms.

The more interesting question is not whether those economics begin to resemble SaaS, but whether AI allows them to become structurally better. Judged through that lens, almost every strategic decision discussed during the analyst day appeared directed towards improving the economics of owning, deploying and operating enterprise networks rather than replacing the underlying infrastructure.

Platform ONE brings together cloud management, AI agents, licensing, workflow automation and multi-vendor visibility into a single commercial platform. Their new enterprise agreements are introducing true-forward consumption pricing. MSP licensing is poolable and AI capabilities are arriving as part of the platform rather than discrete product upgrades.

That all adds up to customers increasingly buying an ongoing operational capability rather than simply another piece of networking hardware. In other words, the commercial relationship is progressively shifting from a transactional hardware purchase to an ongoing software and operational platform. It feels much less like a product strategy than a financial one.

The implication is that the hardware sale increasingly becomes the beginning of a recurring software relationship rather than the end of a product transaction. But that relationship only works if the software is sufficiently capable to simplify, automate and continually improve the operation of the hardware throughout its lifecycle.

If Extreme can continue to execute over the coming quarters, the distinction between its hardware and software businesses may become increasingly irrelevant. The two would become strategically symbiotic. Great hardware creating the opportunity for a long-term software relationship, while great software continually increases the value of the installed hardware. Each reinforcing the other.

This is where the changing measures of AI success become really important for them. AI only matters if it materially improves the operating economics of their ecosystem. Does it increase software attach rates? Does it accelerate deployment for customers and partners? Does it reduce the effort required to support and operate the network? Does it shorten implementation times? Does it create opportunities for greater customer expansion? Does it enable organisations to manage larger, more complex environments without adding headcount?

Perhaps that's the most interesting observation from the analyst day. Platform ONE appears to have been designed not simply to make networks more intelligent, but to make them progressively more economical to own and operate.

For buyers and partners these are not simply product metrics. They are the operational capabilities that will increasingly drive financial performance. They represent acquisition efficiency, onboarding efficiency, implementation efficiency, support efficiency and customer success efficiency. They are, in effect, the new margin economics of AI and they apply equally to networking.

Viewed through that lens, Platform ONE is attempting to do more than add AI to networking. It is attempting to improve the economics of every stage of the customer lifecycle for themselves, their partners and their customers. One observation from the analyst day particularly reinforced this thinking.

Almost every demonstration focused more on reducing operational effort rather than increasing technical capability. They shared a common objective of making enterprise networks easier and less expensive to deploy, operate and support. How refreshing. That means none of the capabilities inherently exist to sell more networking hardware. They exist to increase the lifetime value of the hardware that has already been sold. And that is a fundamentally different economic proposition, which brings us to why this matters now.

Across the world, organisations are entering another major campus networking refresh cycle. Many of the switches and wireless networks deployed during the last decade are approaching end of life. On the surface, buyers appear to be making familiar decisions about speeds and feeds, Wi-Fi 7, switching capacity and security as they always will. But I suspect something more significant is also happening.

Increasingly, organisations will transition from simply selecting networking hardware to choosing the operational model that will govern that infrastructure for the next decade. The quality of the hardware remains essential, but will no longer be sufficient on its own. As a result buyers will start asking more of the type of questions where Extreme has a real competitive advantage, for example:

• Which platform will let my team deploy new sites faster?

• Which platform reduces configuration drift?

• Which one helps me manage more infrastructure without adding engineers

• Which one automates repetitive operational tasks?

• Which one integrates cleanly with the rest of my IT operations?

• Which one continues to improve after I’ve bought the hardware?

One thing seems increasingly certain. Networking remains the operational backbone of the modern enterprise and will not be left untouched by AI. At the same time, buyers face increasing pressure from constrained skills, geopolitical uncertainty, supply chain resilience and relentless expectations around uptime. These forces make the economics of operating the network as important as the technology itself.

So, combining resilient infrastructure with software that continually improves the economics of owning and operating that infrastructure over its lifetime feels less like a product strategy to me and more like the right direction of enterprise networking.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic consulting and advisory services.

Learn more: councilio.co

I've been having a number of conversations lately with equity analysts and fund managers to better understand what I think is AI's emerging coherence problem for the software industry. There is an increasingly visible disconnect between how software companies are being asked to perform, how they are being valued when they do, and how employees are expected to interpret both. This is not simply a post-2021 valuation correction. Something deeper appears to be happening.

The market is no longer asking only whether a company can grow efficiently. It is asking whether that growth is durable, defensible and structurally advantaged in an AI-repriced world. Yet while equity markets appear to have adopted a new set of expectations, those expectations have not been clearly articulated in as simple a way as R40. So if the benchmark has changed, what is it? That’s what I’ve been wanting to know.

For years, the industry has had a reasonably shared language. We all agreed that growth and profitability mattered. The Rule of 40 became a useful shorthand because it gave boards, executives, investors and employees a way to understand whether a software business was balancing ambition with discipline. Then came Rule of 50, and now even Rule of 60 for the stratospheric performers. But now that language feels insufficient.

Because the problem is not that software companies have stopped executing. It is that the financial market appears to have moved beyond the benchmarks by which that execution has traditionally been measured. This means that companies delivering Rule of 40-plus performance are still attracting significant valuation discounts, while others with comparatively weaker operating metrics appear to be judged against an entirely different set of expectations.

Today a software vendor can deliver strong revenue growth, improve free cash flow, secure high renewal rates and a cleaner cost base, and still take a big hit. The issue is not deteriorating performance, but uncertainty about how AI will reshape the company's future economics.

We have reached the point where if investors cannot confidently assess whether AI will compress pricing, erode product differentiation, shift value towards infrastructure, weaken seat-based business models or fundamentally reshape the economics of software development, go-to-market and customer delivery, they are likely to price that uncertainty into today’s valuation. That has become the default.

Their question is no longer simply, “Is this a Rule of 40 company?”. It is, “What kind of Rule of 40 company is this?” More importantly, what additional characteristics determine whether the market assigns a premium or a discount? Part of that uncertainty stems from the fact that we are still learning where AI genuinely changes enterprise software economics and where it does not.

AI has become remarkably capable of generating components, prototypes and functional pieces of software. Anyone who has experimented with frontier models has seen how quickly a feature, workflow or user interface can be produced through vibe coding or agentic development. But building an enterprise platform is not the same as building a feature.

Enterprise software must still satisfy architectural standards, cybersecurity requirements, governance controls, data models, integration patterns, operational resilience, compliance obligations and long-term maintainability. Generating an isolated capability is one thing. Delivering an entire platform that consistently meets those standards is another.

But until we have stronger evidence about where AI genuinely changes those economics, markets will continue to price future value through uncertainty rather than assumption. Those valuations then become today's management decisions. And that is where I think we need to momentarily leave the financial markets and consider how future expectations are already reshaping the software workforce.

Share

This is a tough time to be inside tech. Inside many software companies, employees are being asked to do two psychologically conflicting things at once. They are being asked to maintain belief, energy and cultural enthusiasm while watching the equity market seemingly discount the very operating performance they were told to deliver. At the same time, they are being told to upskill in AI because AI will be critical to their future employability, while also seeing AI used as the rationale for restructuring, hiring freezes or role elimination.

We have moved past hype to the point where that is no longer a small cultural issue. It is fundamentally breaking the emotional contract between company and employee. Executives can continue to tell staff that the company is executing well. They can point to operating discipline, product innovation, customer wins and AI transformation. But if employees see those gains fail to translate into equity value, compensation upside, job security or organisational confidence, the message becomes harder to sustain. Performance without reward eventually becomes fatigue. And when future uncertainty begins driving today's restructuring decisions, fatigue quickly becomes fear. I hear about redundancies every day now.

In conversations this year, I've also sensed that fatigue beginning to emerge among the leaders of some of the world's largest software companies. Many have spent years executing difficult transformations, improving operations and reshaping their businesses for the AI era, yet still find themselves explaining why sustained execution has not translated into the market recognition they expected. Their challenge is not simply one of execution. It is that the market now appears to be judging them against a benchmark that has never been clearly articulated.

This feels like the point where equity and financial analysts may need to do more than simply apply lower valuation multiples and wait for management teams to infer the new rules. If the market has changed the benchmark, it should be stated more clearly. That creates accountability. It also allows boards, management teams, employees and investors to understand what success now looks like and to judge performance against a shared set of expectations rather than an implied one.

That single shift changes almost everything. The market appears to be supplementing growth and profitability with a broader assessment of the operational characteristics that determine whether those economics are sustainable. In practical terms, the emerging benchmark seems to evaluate software companies on the quality of their growth, the durability of their economics, the defensibility of their AI capabilities and, ultimately, their ability to create better long-term customer outcomes rather than simply generate short-term sales activity.

Those characteristics cannot be observed directly through traditional financial reporting, so investors will have to increasingly look for operational evidence.

They still want to understand where profitability is coming from, whether customers continue to expand their investment over time, and whether pricing power can be sustained. But now also whether AI capabilities create genuine competitive advantage, and whether gross margins will remain resilient once AI infrastructure costs are fully accounted.

Increasingly, they are also paying attention to measures like sales efficiency, implementation velocity, and product-led adoption that were once considered secondary. We could throw a blanket over all these and just consider whether customers are able to achieve better outcomes with less friction. This remains one of the biggest unanswered questions in software today.

This is where the work of industry analysts has always differed. Beyond the polished demonstration and the marketing campaign sits the much harder questions of execution. How quickly can customers adopt the technology? How successfully can they implement it? And do they ultimately achieve the outcomes that were promised? Helping organisations answer those questions has long been one of the defining roles of industry analysts. In doing so, we have spent decades evaluating the operational realities that sit between a vendor's promise and a customer's outcome. Those are now becoming some of the defining questions for investors as well.

Put another way, the two disciplines of equity and industry analysts are beginning to converge. Equity analysts largely explain what happened to a company's economics through its reported financial performance. Industry analysts have traditionally explained why those economics happened by observing product maturity, implementation reality, customer adoption and competitive execution.

As software valuation shifts more towards operational capability rather than purely financial efficiency, those two perspectives become increasingly complementary. The reason is that the market is beginning to distinguish between companies that reduce costs by genuinely improving the business through AI, and those that reduce costs by simply redistributing effort, complexity or risk elsewhere in the value chain. That distinction has long sat at the heart of industry analysis. We have always been interested not simply in whether a product could be sold, but whether customers could successfully implement it, adopt it and realise value from it.

In the AI era, productive efficiency is becoming more valuable than simple efficiency. The market is no longer asking only how efficiently software companies generate revenue. It is increasingly asking whether AI is removing friction from the system or merely moving it. Ultimately, that means asking how efficiently software companies create customer success.

That distinction becomes much easier to understand through a practical example. One emerging assumption in software markets is that reducing the cost of sale will become an important part of the new equity story. On a spreadsheet, or a boardroom whiteboard, that makes perfect sense. If AI can reduce the cost of demos, prototypes, solution design and technical validation, software companies should be able to lower sales and marketing expense, improve margins and shorten sales cycles. Job done. And to be fair, I have seen some genuinely impressive applications of AI in this area. But by the same token, we’re yet to answer the most important question about what happens to those productivity gains?

So where have I landed after all my discussions? I think the equity analysts are absolutely right to ask harder questions. The old valuation models did not fully account for AI disruption, seat compression, infrastructure cost, product commoditisation or the growing fragility of software moats.

I think industry analysts are equally right to focus on customer outcomes, implementation reality, organisational readiness and the distinction between capability and adoption. Demonstrating that AI can do something has become relatively easy. Demonstrating that customers can consistently adopt it, govern it and realise value from it at scale is considerably harder and will be beyond many protagonists.

And I think operators are right to pursue efficiency. But efficiency only creates long-term value if it improves the business rather than simply shifting effort, complexity or risk somewhere else in the customer lifecycle.

Each perspective is valid. The problem is that they are still being assessed through different lenses. I return to the missing piece being a shared benchmark that connects financial performance, operational execution and customer outcomes.

The next software valuation model can’t simply reward lower cost of sale or lower cost to serve, though these are very important. It should reward productive efficiency via the ability to grow, expand, implement and retain customers with less friction and better outcomes. And this is where the industry analyst perspective becomes critical.

We see how a vendor’s sales pitch aligns with customer implementation reality. We hear whether products are easier to adopt, whether customers are reaching value faster, whether partners are carrying more delivery risk, and whether support demand is falling because the product is better or because support has been reduced.

Reduced cost of sale should therefore not be treated as inherently positive without further interrogation. Are customers going live faster? Are they expanding faster? Are support needs falling because the product has improved, or because support has been cut? Are SE reductions improving scalability, or merely pushing technical risk downstream into delivery, partners and customer success?

The real question is not whether the vendor, through AI, has become cheaper to operate. It is whether the vendor’s solution, through AI, has become easier to buy from, easier to implement, easier to adopt, easier to service, and easier to expand.

So the market is right to move beyond Rule of 40. The challenge is that the characteristics increasingly determining software value are not readily observable through public financial reporting.

Rule of 40 democratised software valuation because anyone with a company's published financial statements could calculate it. The emerging AI benchmark is fundamentally different. Many of the characteristics now driving premium or discounted valuations are operational rather than financial. They are often difficult to observe from outside the business, forcing investors to infer them. And when markets cannot directly observe what matters most, they tend to price uncertainty conservatively.

So until investors can better observe those drivers, markets will continue relying on imperfect financial proxies. Companies will manage to largely unstated expectations, and employees will struggle to understand why strong execution is not translating into value creation. But perhaps, we can say with most certainty, Rule of 40 gave the software industry a common language, whereas the AI transition has exposed the need for a new one.

And until we become better at observing the operational characteristics that now determine software value, investors will continue pricing uncertainty, management teams will continue executing to largely unstated expectations, employees will continue questioning whether performance is being recognised, and customers will (almost certainly) ultimately absorb the hidden cost of margin improvement. The challenge is making what matters visible.

My own contribution is this. Firstly, high growth and strong margins remain necessary. Of course they do. They are simply no longer sufficient.

Secondly, the market appears to be placing greater value on operational capability rather than simply financial efficiency. That naturally raises the new question as to where should that capability become visible?

I think the answer is that every release, every product investment and every AI capability should systematically reduce friction across the customer lifecycle. Is AI making the company easier to buy from, easier to implement, easier to adopt, easier to serve, easier to trust and easier to expand? Those are questions we can already begin to assess, even if we cannot yet measure them consistently from public financial reporting.

That leads to a simple conclusion. If two companies deliver identical Rule of 50 performance, but only one can demonstrate that AI is systematically reducing customer friction without increasing customer risk, they should not command the same valuation. That difference is precisely what the next generation of software valuation needs to capture.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic consulting and advisory services.

Learn more: councilio.co

James eventually joined IBM, where he played a leading role in its software and Lotus Notes businesses, most notably during the formative years of corporate social computing, helping organisations rethink the collaboration, knowledge sharing, and digital work we take for granted today. He later moved into the world of pre AI-powered med-tech, where he helped bring innovative technologies to market that improved safety and human performance. Today, he leads Gorry Advisory, helping organisations navigate the opportunities and challenges of the AI era.

To dive deeper into James’s insights, check out the full conversation and reach out to him at info@gorryadvisory.com.

About Councilio

Councilio provides independent tech analysis, executive briefings, keynote speaking, workshops, consulting and strategic advisory services.

Learn more: councilio.co

For context, I was pondering, like many others, what happens when the AI market experiences its own moment of disruption. Maybe it’s a bubble. Maybe it’s a supply chain shock. Maybe it’s some black swan event we haven’t even imagined yet. Looking back at the architectural patterns I was writing about at the time, it led me to a different question. What matters more for businesses? The biggest model, or the most resilient environment?

History tells us that every major technology boom eventually moves from experimentation to consolidation. The internet survived the dot-com crash. Cloud survived the hype cycle that followed. What mattered in the end was not who grew the fastest, but who became indispensable enough to remain standing when the market corrected.

So if the current AI boom were to experience a similar correction, the most important question may not be which company has the smartest model but which ones can continue delivering AI at all. That distinction matters because of the surprisingly narrow foundation of today’s AI ecosystem.

Despite the appearance of intense competition, an enormous portion of the market ultimately depends on the same underlying compute infrastructure. Models and applications and agents are in fierce competition. But beneath the surface much of the industry converges on the same supply chain.

That makes the AI economy a resource economy and history has not always been kind to industries built upon assumptions of unlimited resource availability. This is where IBM stands out.

Unlike many organisations currently benefiting from the AI boom, IBM’s future appears less dependent on AI enthusiasm itself. That feels like an important distinction, particularly when viewed through the lens of the cloud era. During the early 2010s, IBM competed aggressively in a market that ultimately consolidated around hyperscale cloud providers. It is difficult not to conclude that some important lessons were learned and, as a result, their AI strategy emerging today looks very different.

Even if the current AI boom came to an abrupt end tomorrow, organisations would still need to process transactions, serve customers, move goods and make decisions. A significant portion of that operational reality continues to touch IBM technology somewhere within the stack. That does not make IBM immune to an AI correction. Growth expectations would certainly change and investment priorities would definitiely shift. But IBM would still have a substantial role to play in helping customers apply AI to mission-critical workloads.

So while much of the market has been competing to build smarter AI, IBM appears to have spent equal time preparing for a world in which AI itself becomes a source of instability. The objective is not necessarily to win the AI race. It may be to ensure that customers can continue running it regardless of how the race unfolds.

Their approach seems to be based around the question of exactly how much AI capability organisations need, and where they need it. That’s subtley different from where much of the current market is in assuming that increasingly powerful AI requires increasingly large AI infrastructure.

Red Hat, OpenShift, watsonx, Granite, hybrid cloud, AI governance and physical infrastructure are collectively a very compelling product strategy. But together they form part of something far more profound. Through investments in IBM Z, Power, Telum and Spyre, IBM is effectively asking whether many enterprise AI workloads can be delivered closer to where work already occurs. In tech-lish, that translates to a simple proposition. Not every business problem requires a frontier model trained and operated on dedicated AI infrastructure. Many can be solved closer to where the work, data and transactions already exist.

For example, fraud detection, decision support, workflow automation and operational intelligence may prove just as valuable when delivered directly within existing enterprise environments, closer to the transaction, the process and the decision itself. In other words, rather than moving the work to the AI, move the AI to where the work already happens. And perhaps, welcome hybrid back to the AI discussion.

This is perhaps the most overlooked aspect of the current AI debate. Most organisations do not exist to train AI models. AI is not the purpose of the business. They exist to process transactions, serve customers, manage assets and make decisions. Historically, enterprise computing has been optimised around those outcomes, measured by things like reliability, throughput, availability, security and predictable operating costs. AI introduces a new capability layer, but it does not eliminate those underlying requirements. In fact, it may reinforce them.

I’ve written elsewhere about the cost problem, most recently here, and it’s a topic we’ll keep circling back to until the industry resolves it. Because no matter how impressive the model, it ultimately has to justify itself against the economics of the business process it is intended to improve.

For now, the market has largely avoided answering the more fundamental question as to whether the future of enterprise AI belongs exclusively to organisations operating enormous GPU clusters, or whether a significant proportion of business value be generated much closer to the transaction itself. During this boom that we are in, that distinction barely matters. Infrastructure is plentiful, investment is outrageously abundant and organisations naturally gravitate towards the most capable solutions available.

HOWEVER, a major market correction immediately changes the equation. Suddenly the question is no longer theoretical. It becomes one of the first questions the market explicitly or implicitly answers. Organisations will gravitate towards the architectures that continue to function under constraint (the 21st century constraint we have to reinvent for).

Telum is a useful example because it embodies a very different architectural assumption. Rather than moving the transaction to the AI, what if the AI happens where the transaction already exists? Today that distinction barely matters because AI infrastructure is plentiful and the market naturally gravitates towards capability in the form of bigger models, faster inferences and larger clusters.

But disruptions have a habit of exposing assumptions. When infrastructure becomes constrained, expensive or unavailable, the question changes from who can deliver the most AI to who can continue delivering enough AI to keep the business running. That is where IBM’s architecture rises to the top.

Unlike many participants in the current AI ecosystem, they are not dependent on a single layer of the stack. Post Cloud 1.0 IBM has spent years assembling an end-to-end architecture capable of delivering AI across multiple environments. I am not for a second suggesting that they can replace the hyperscale AI ecosystem. But I am certainly suggesting that IBM possesses enough of its own infrastructure and AI capability to continue creating meaningful business value while others are still trying to secure access to theirs.

If that assumption proves correct, resilience may not simply come from owning infrastructure but from owning enough of your own infrastructure to keep operating when the market around you becomes unstable.

The concept of IBM’s AI resilience does not stop with IBM. It extends to every software provider, partner and customer building on top of that ecosystem. Because, especially in Australia’s mid market, infrastructure resilience is rarely something organisations buy directly. It is something they inherit, and often without even realising it. So the Australian organisations most insulated from an AI infrastructure correction may not be the ones operating AI platforms themselves. They may be the organisations consuming software and services built upon platforms that were designed to endure disruption in the first place, even if they weren’t the ones to think about it.

Consider Australian software provider Pronto Software, and the thousands of mid-market organisations that depend on them every day. Most customers think they are purchasing software, when what they are really purchasing is software, architecture and resilience.

So when a software provider, like Pronto, builds AI capabilities on top of an enterprise-grade ecosystem, their customers gain more than access to new functionality. They gain access to the resilience characteristics of the underlying platform itself. And in these uncertain times, even beyond AI sovereignty, that will prove increasingly important.

A Pronto mid-market manufacturer, distributor or local government customer is unlikely to care which model generated an answer. Nor whether a recommendation or report originated from one foundation model or another. They care whether payroll runs and whether orders are processed. They care whether customers are served and whether operations continue.

In that environment the discussion shifts from AI capability to AI continuity. And they are not the same thing. Because when the excitement fades, and we all remember how quickly it does, organisations will discover that intelligence was only one part of the equation. The real test was whether the architecture could endure, whether the platform could continue operating, and whether their tech partners showed up for work on Monday.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

I had a lot of conversations last month with regional MSP and ERP leaders. We largely spoke about all things AI, software and hyperscale platforms, general customer demand and the future of their respective industries. Much of the discussions covered ground that would be familiar to anyone working in technology right now. Data readiness is becoming more important. Customers are remaining trapped somewhere between AI workshops and production deployment. And automation, once considered yesterday's story, is quietly becoming fashionable again. None of that surprised me.

What did catch my attention though was the point at which, in conversation after conversation, with execs who had never spoken to one another, every discussion drifted into stories about their efforts to build intellectual property from the knowledge they have spent years accumulating.

This is not a new idea. The industry has long recognised the value of capturing knowledge and reusing it across future customers. Methodologies, accelerators, industry templates and centres of excellence all represent attempts to turn experience into something more durable than an individual project. We can see elements of this thinking in TechnologyOne's SaaS+ model, ServiceNow's implementation accelerators, Accenture's industry assets, IBM's industry frameworks, Infosys Topaz, Cognizant Neuro and countless other examples.

The challenge has never been recognising the opportunity. The challenge has been the cost and effort required to systematically capture, refine, maintain and scale that knowledge. As a result, much of the value created during customer engagements has remained trapped inside individuals, project teams and specific implementations. The opportunity has always existed. What is changing is the economics of realising it.

My recent conversations across the market increasingly point to two realities. First, most firms now recognise that they need to pivot (again) to solve this problem once and for all. Second, far fewer are clear on what they should pivot towards and are still searching for a sustainable source of competitive advantage. Those that are clear are increasingly focused on creating intellectual property. Products.

This is where the distinction becomes important. Many firms have spent years attempting to package expertise through methodologies, accelerators, frameworks, templates and more recently managed services and SaaS+ offerings. These approaches have undoubtedly improved delivery efficiency and helped organisations scale. But the product still exists to make the service more efficient. It does not yet replace the service as the primary vehicle for value creation. The underlying economics remain rooted in labour. What is emerging now is a model where expertise itself becomes the asset, and increasingly, the product.

So the next step is fundamentally different. Rather than simply making services easier to deliver, it seeks to transform expertise itself into a reusable asset. An asset that can be deployed, refined and scaled repeatedly without requiring a proportional increase in people, projects or delivery effort. Or cost to the customer. In that model, knowledge does not merely support the service. Increasingly, it becomes the product.

Share

For much of the past thirty years, the managed services industry has operated according to the relatively simple formula of revenue as a function of people. More customers required more consultants. More projects required more project managers. More support contracts required more support staff. Growth was linear and predictable because delivery capacity was ultimately constrained by the number of people available to perform the work.

That model produced some extraordinarily successful businesses, but it also created a structural weakness whereby knowledge was constantly being generated, yet very little of it accumulated. Most of the value remained trapped inside people rather than becoming a durable asset of the delivery organisation. For years this was largely ignored because demand for technology services continued to grow. Today, however, financial markets are beginning to expose the limitation.

Across the NYSE, Nasdaq and India's major exchanges, investors are increasingly distinguishing between services businesses that scale through people and those that scale through intellectual property. The result is a growing valuation gap between organisations whose future depends primarily on hiring more staff or acquiring the next $10 million growth firm for functional or regional expansion, and those capable of systematically transforming expertise into reusable assets, software, automation and platforms. The shift is subtle but important. Growth is becoming less about accumulating capability and more about industrialising it. The firms creating the most value are increasingly those that can capture what they already know, refine it and apply it repeatedly at scale.

Lets’ look at a couple of common examples. A project team might solve a complex integration problem for a manufacturing client. Or a consultant would develop a highly effective process improvement approach for a council, state or federal government. Maybe a support engineer discovers a faster way to resolve recurring incidents. While the MSP organisation benefited from those experiences, it was often only indirectly because much of the expertise remained trapped within individuals and teams. Knowledge was constantly being generated, but much of its value walked out the door at the end of the day.

Most discussions about AI today still begin and end with labour. The focus on productivity, efficiency and headcount is understandable, but it is also causing us to overlook something far more significant. The more practical, and therefore interesting, question is what happens when contextual expertise becomes easier to package and reuse. What happens is that knowledge that previously disappeared at the end of a project can be transformed into IP that compounds over time. And that’s the lens through which MSPs will be judged. It means where the MSP of 2015 sold effort, the MSP leders of 2030 will sell accumulated specialised intelligence.

Evidence of this shift is already beginning to emerge. Recent moves by Coforge in the insurance sector demonstrate how technology service providers are increasingly combining domain expertise, proprietary intellectual property and AI capabilities into highly specialised industry offerings. The company is not positioning AI as a generic productivity tool. Instead, it is applying agentic AI directly to underwriting, claims management and insurance operations where years of accumulated industry knowledge create a genuine competitive advantage.

What connects examples like Coforge and other MSPs is not artificial intelligence itself. It is the recognition that expertise can be transformed into new informaiton assets. For decades, service providers primarily monetised knowledge through labour. They sold projects, consultants, developers and support teams. Increasingly, however, they are needing to find ways to monetise that same knowledge through software, workflows, automation, platforms and intellectual property. AI is not creating this trend but it is first unlocking and then accelerating it.

There is a reason it hasn’t happened before now. Historically, transforming expertise into a reusable asset required significant investment. Product development teams were expensive. Building software with the appropriate controls and governance often took years. Documentation was labour intensive and difficult to maintain. Capturing, refining and codifying organisational knowledge was not only slow and cumbersome, it was rarely the primary objective. It was simply a by-product of delivering projects and serving customers. In many cases, the effort required to formalise that knowledge worked against margins rather than enhancing them. As a result, much of the value remained embedded within people and projects instead of becoming a durable asset of the business itself. That is the friction that is now beginning to disappear.

A managed services provider that solves a problem once can increasingly convert that solution into something repeatable, whether that be through a reusable integration or turning an automation into a functional agent. By the time the tenth customer arrives, the organisation is no longer selling the same thing it sold to the first and what began as a consulting engagement has quietly evolved into intellectual property.

This is why the direction being taken by companies such as Coforge matter. Customers will not ultimately buy artificial intelligence but they will buy better underwriting, faster claims processing, improved risk assessment and more effective operational outcomes. So as artificial intelligence becomes ubiquitous, deep industry expertise will still be hard won, for now. Therefore as AI lowers the cost of creation, the value of specialised knowledge will only increase rather than diminish.

This may explain why some of the most successful managed services businesses are becoming increasingly specialised. They are not looking to narrow their horizons because they lack ambition. They are doing so because specialisation creates defensibility and while a competitor can easily replicate a technology, replicating twenty years of accumulated industry understanding is considerably harder.

One of the things this means is that the companies best positioned for the AI era may in fact be service providers that learn how to behave like software companies with vertical specialisations. Not by abandoning services or replacing people, but by transforming expertise into assets and amplifying the value of what their people know.

So overall I think the future of the managed services industry looks pretty strong, albeit confused as hell in 2026. The sector will evolve into something that sits somewhere between consulting, software and platform businesses. The winners will still employ talented people, and those people will still provide advice, and they will still solve difficult problems. The difference is that every problem solved will create another asset and every engagement will strengthen a growing bank of intellectual property. Those with strong PaaS partnerships will do better than those stuck in ERP land alone.

More importantly, that intellectual property will not be generic. It will increasingly reflect the accumulated experience of particular industries, customer segments and operational environments. The value will not come from building another workflow, integration or automation. It will come from building or deploying assets that embody years of hard-earned domain expertise.

We can already see this pattern emerging in other industries. Platforms such as Lovable are not simply automating web development. They are packaging decades of accumulated design, engineering and product development knowledge into a reusable asset that can be applied repeatedly across millions of projects. The customer is not really buying AI, nor an army of developers. They are consuming expertise through a different delivery model.

The same principle increasingly applies to MSPs. Over time, accumulated industry knowledge, delivery experience and operational understanding will behave less like labour and more like capital.

The conversations I’m having across the market increasingly point in the same direction. Most firms know they need to pivot and the ones that appear furthest along are building intellectual property. That means where the MSP or servies firm of 2015 sold effort, the MSP of 2030 will still sell expertise, but increasingly it will package, scale and monetise what it learns from the projects. That is step one.

The more difficult question is whether traditional services firms are equipped to make that transition. Because building a successful services business and building a successful product business require very different capabilities. One optimises for customer-specific outcomes. The other optimises for repeatability and scale. One rewards flexibility. The other rewards standardisation.

Every firm will ultimately embrace AI. It's just that some will use it to work faster inside the existing factory, while others will use it to rebuild the factory itself. Only one of those will make selling hours less important.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

During his time, Colin helped dramatically elevate ServiceNow’s market presence and strategic relevance. The “Put AI to Work” campaign has successfully repositioned the company into the centre of the enterprise AI conversation, while the collaboration with brand ambassador Idris Elba has helped humanise and amplify that message at global scale.

So ServiceNow has largely won the first battle. The market now understands the company matters, technologists are increasingly recognising the broader architectural direction toward PaaS and enterprise orchestration, and investors are now trying to comprehend the scale of the opportunity sitting in front of the business.

But awareness is not the same as understanding and I think a translation challenge remains. That means more work is required to help customers, partners and even internal teams understand where to begin, how the platform fits together, and how enterprises practically move from today’s operational complexity toward their own AI-enabled futures.

ServiceNow has evolved beyond the category leader that originally made it successful, but not yet escaped the gravitational pull of the category itself. I don’t see that as a bad thing. No serious technology executive now sees it as simply an ITSM or ticketing platform. But neither is it easily understood en masse as purely a workflow company, a CRM vendor, or even an AI story. Their strategic expansion has become both a great strength and their greatest communication challenge.

As enterprises move toward an agentic future, ServiceNow definitely has a strong claim to the emerging operational execution layer category. The continuing difficulty is that enterprise buying structures still reflect the fragmented technology models of the past. Procurement remains dispersed across disconnected buying centres, while customers continue evaluating technology through narrower category lenses that no longer fully reflect how modern platforms operate, i.e. the ServiceNow vision. The inheritance for the next CMO is therefore not an awareness problem, but a translation problem. Here are a couple of examples.

Firstly, ServiceNow appears to understand their larger destination clearly. The challenge is that it has heavily weighted the market narrative toward the scale of the opportunity, without an equally aggressive focus on maintaining the base and winning the decomposed competitive battles occurring at the individual layers of enterprise technology where customers are continuing to make buying decisions.

By that I mean, each of ServiceNow’s major offerings is strategically valid. Each addresses a real market problem, competes credibly with category incumbents and has the potential to generate meaningful long-term growth. But some of them are getting lost in the noise. The challenge is not the logic of the portfolio itself. It is the operationalisation of the story surrounding it.

ServiceNow has spent several years building a more unified NOW Platform narrative, and the Agentic Blueprint presented at Knowledge 2026 was an important step forward. It helped structue an increasingly broad platform into clearer conceptual categories.

But simplifying the platform is not the same as simplifying the customer journey into the platform. Enterprise transformation, to this point, rarely begins with a platform strategy. Organisations enter through immediate operational pain points. For one enterprise, that may still be ITSM modernisation. For another, it may be security operations, customer workflows, identity governance, regulated industry processes or increasingly CPQ, as disruption reshapes traditional CRM markets.

While these workloads solve very different operational problems, they are ultimately just entry points into a much larger opportunity sitting beneath the enterprise itself. The next marketing challenge may therefore be less about product positioning and more about journey orchestration. That requires more of a focus on helping organisations understand not just where to begin, but how one operational challenge progressively connects to a broader AI-enabled execution model over time.

Secondly, from a buyer perspective, the company is now competing with almost everyone. That was unclear just a few years ago. Not today. Hyperscalers are directly entering executive AI conversations, often with a cleaner path. Identity vendors are experiencing a resurgence driven by agentic AI and machine identities. Traditional workflow and BPM vendors remain deeply embedded in regulated industries. CRM is obvious. And frontier AI companies are reshaping customer expectations faster than enterprise software sales cycles can realistically adapt. Now security. That creates enormous pressure across the broader ServiceNow ecosystem.

Internal sellers are increasingly expected to navigate highly specialised domains that cannot simply be collapsed into a single platform conversation. CRM cannot be sold like ITSM. Security is a very difficult sell for generalists. CPQ may generate short-term growth opportunities, but it is probably too specialised to become the primary pathway through which most enterprises discover and expand into ServiceNow’s broader operational execution model.

Partners are simultaneously being pushed to evolve from implementation channels into transformation orchestrators, and resisting. Customers, meanwhile, are being asked to absorb and operationalise a pace of innovation arriving faster than most enterprise digestion cycles naturally allow. Yet, as we saw at K26, the market narrative continues accelerating regardless. It’s exhausting.

The challenge for ServiceNow is therefore no longer simply innovation. It is more about selectively, not collectively, re-aligning product ambition with sales capability, partner maturity and customer readiness at a time when the entire enterprise technology market is converging faster than its traditional operating models were designed to support.

Thirdly, they could rethink what constitutes a valuable customer reference in the AI era. Organisations such as Orica and Standard Chartered are undoubtedly important transformation stories. But they have also become ubiquitous across the broader technology ecosystem, especially in Asia Pacific. The same organisations now appear repeatedly every year across hyperscaler, AI, security, platform and consulting conference narratives, often simultaneously collaborating and competing with one another in the same customer environment.

At some point, these references stop feeling like differentiated customer stories and start feeling like isolated ecosystem case studies. The challenge is that most businesses are nowhere near that level of transformation maturity, operational scale or investment sophistication. What many normal customers increasingly need are more relatable transformation journeys. Organisations still navigating fragmented systems, regulatory constraints, operational debt and incremental progress toward AI-enabled execution. (Like the State of Hawaii guy from Knowledge).

In a market still trying to understand how AI operationalises inside the business, authenticity, trajectory and practical progression may now resonate more strongly than another perfectly polished North Star example.

Fourthly, most enterprises are now carrying decades of accumulated operational complexity in the form of duplicated processes, disconnected systems, fragmented approvals, manual interventions and regulatory overlays. These thousands of small inefficiencies have compounded over years of technology expansion and are especially prevalent in heavily regulated industries where platforms such as Pegasystems and Appian have traditionally thrived managing process-intensive operational environments.

This kind of operational complexity has traditionally been framed purely as a problem statements ranging from bad technical debt, to bureaucratic burden (red tape), and legacy process to be removed through large-scale transformation programs. But I’ve always felt that view is way too blunt and misses something important.

Much of this complexity represents years of accumulated institutional knowledge, embedded operational logic and regulatory adaptation that the right people in the right room can still unlock. In many ways, it resembles crude oil. It’s unrefined and difficult to work with yet still immensely valuable once properly extracted, analysed and refined. The problem has never been that this complexity existed. The problem was understanding how to surface it, structure it and translate it into operational value.

This is where their capabilities in process mining become strategically important. Not simply as observability tools, but as discovery engines capable of exposing how work actually happens across fragmented systems and disconnected teams. They can surface the invisible operational friction that traditional transformation programs often fail to identify. In that framing, Agentic AI is not actually the starting point. It is the starting point of refined complexity.

This is an area that makes sense, to they and their partners, as to their greatest long-term advantage. The company performs best in environments where complexity, regulation and organisational fragmentation already exist. And those are precisely the environments where hyperscaler-only approaches and isolated AI tooling often struggle to operationalise meaningful outcomes.

So I think the next CMO inherits a role that looks less like traditional enterprise software marketing and more like enterprise systems diplomacy. The task is no longer simply generating demand. It is helping organisations understand where to begin, how to progress, and why the operational complexity already sitting inside their business may actually contain the blueprint for their AI future.

That will require a structural evolution in go-to-market strategy itself, both globally and regionally, involving greater sales specialisation, stronger industry-led motions, more relatable customer references and clearer transformation pathways. It cannot achieve its full potential by trying to sell everything everywhere all at once. It is more likely to win by helping organisations move from one solvable operational problem to the next logical stage of maturity.

The irony of course is that ServiceNow sells execution, yet its own next phase depends on whether it can execute its strategy with enough focus, structure and discipline to carry the market with it. Because the company is not wrong about where enterprise technology is heading. They are absolutely right. It’s just that in many ways, it is earlier than much of the market in recognising that the future will be defined less by standalone applications, and more by orchestration of operational decision-making across fragmented environments. As an example, see my recent post on IBM to see what this means for Security.

But seeing the future and operationalising it are two very different things. They do deserve success. For that to happen, the vision now has to become consumable for customers, repeatable for sellers, and believable for partners.

And finally, I can’t wait to see what Colin helps shape next at OpenAI. But leaving the industry with Idris Elba performing at the ServiceNow Knowledge 2026 afterparty at The Sphere in Las Vegas feels like a fitting final scene.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

The questions themselves were straightforward. But the answers, and more importantly the implications behind them, did not merely hint at IBM’s strategy or provide insight into where the market is heading. They exposed it directly and then walked us through the front door.

First, the questions.

1. Why did IBM choose to remain in the Managed Security Services market, particularly when it had exited so many other infrastructure-heavy areas of global tech services? The question was of interest to me because IBM’s security services have appeared to be shuffled in recent years, as the company attempted to find the right home within its overall strategic changes. In the process creating a degree of frustration and uncertainty for staff and customers about future fit. That’s a commercially important consideration.

2. Were the economics of agentic AI fundamentally reshaping cyber warfare? I was interested to hear how AI was impacting large-scale, resource-intensive attacks like DDoS. Was IBM seeing them become less attractive as a threat vector, while simultaneously observing an increase in the viability of lower-cost, higher-impact threats such as phishing, disinformation, supply chain compromise, and data leakage? Also, a commercially important consideration for cyber investment.

What emerged from those discussions cuts to the heart of changes unfolding across the global security, AI and managed services markets. Perhaps you’ve wondered yourself. What does actually change when agentic systems begin making decisions, not just detecting events? And if that happens, what does “managed services” even mean anymore? IBM has the answer.

Six years into his tenure as IBM CEO, there is still a version of the Arvind Krishna IBM transformation story that gets told as a retreat. It is the one where they lost to the Indian SI and Managed Services conglomerates. Where they exited infrastructure-heavy tech services, carved out large parts of thier business into a $20 billion start up called Kyndryl in 2021, stepped away from the global outsourcing market they helped define, and left customers wondering how they had misread another cyclical change.

Yet at the same time, it didn’t read as a clean break. IBM chose to retain and continue investing heavily in cyber security, an intensely infrastructure-heavy function. Viewed through the traditional lens of managed services and systems integration, that carried all the signs of an inconsistent and confused message. It was the basis of my first question.

But over time, something IBM is famously consistent in drawing on, the inconsistency began to disappear. The issue was not necessarily IBM’s strategy. It was that many of us were still applying an older definition of managed services to a market that was already shifting underneath it.

If managed services were still primarily about operating infrastructure and running environments, IBM’s security strategy was difficult to reconcile, even for its own staff. But if the future of managed services is increasingly about managing decisions, orchestrating responses, and governing execution in AI-driven environments, the strategy makes far more sense.

In hindsight, it may simply have taken five years for the strategy to become legible. Quietly, and with relatively little fanfare, IBM shifted toward managing cyber decision-making at scale rather than simply operating security infrastructure. As a proof point, IBM is not alone in moving this way, and that transition is increasingly visible across the industry today. Step outside raw infrastructure and ServiceNow’s recent direction, product positioning, and acquisitions arguably represent the software platform equivalent of the same broader transition.

The traditional model of managed services was built around stability and predictability. Hundreds of billions of dollars of contracts have been written on the expectation that systems would be kept running, service levels would be met, and issues would be resolved within defined timeframes. It was labour-intensive, operationally focused, and increasingly constrained by cost pressures as infrastructure became more standardised, more easily commoditised, and labour arbitrage margins sent into retreat. For decades, value creation has been tied to the idea that these environments required constant human oversight to function effectively. So what changed?

Apart from AI itself, cyber may be one of the clearest examples of why traditional managed services thinking is beginning to break down. While security services historically behaved relatively predictably for decades, cyber does not.

It is not a steady-state environment that can simply be optimised around predictable workloads, standardised controls, and defined response times. It is a continuous exercise in judgement under uncertainty, where signals are incomplete, context shifts rapidly, and the consequences of getting decisions wrong are highly asymmetric. A single missed event can carry more weight than thousands of correctly handled ones.

The speed at which those judgements now need to be made is also increasingly incompatible with human response times alone. That changes the nature of the service entirely. The centre of gravity shifts away from merely operating infrastructure toward orchestrating detection, interpretation, prioritisation, and coordinated response across constantly changing environments.

This is where agentic systems begin to have real impact, because they operate naturally in environments where decisions need to be made continuously, at speed, and with high levels of confidence. Going back as far as the late 2000’s IBM used to coin this as the ability manage the three Vs. Volume. Velocity and Veracity.

That sets the basis of the change. But the shift becomes even more significant when viewed from the perspective of the attacker, because the same forces transforming cyber defence are also transforming offensive capability. That was the basis of my second question.

Not all forms of attack will benefit equally from this change. Blunt-force methods such as distributed denial of service (DDOS) still rely on scale, coordination, and infrastructure, and while they remain relevant, are increasingly well defended by the large platforms that sit at the edge of most enterprise environments. Agentic systems do not fundamentally change that equation, because the problem remains one of volume rather than understanding.

Other attack vectors, however, begin to look very different and increasingly emerge as priority focus areas for cyber security strategies in the age of agentic AI.

Phishing becomes more targeted and more convincing as systems are able to build detailed contextual profiles and adapt messaging in real time. Data exfiltration becomes quieter and more deliberate, focusing less on volume and more on identifying and extracting high-value information without triggering detection. Supply chain attacks become more attractive because of the incredible leverage they offer, allowing a single point of compromise to cascade across multiple organisations, and expontentially erode brand trust and its adjacent financial blast radius. Even disinformation becomes easier to generate, easier to personalise, and harder to attribute.

What links these threats together is that they rely less on brute force and more on understanding, adaptation, timing, precision, and increasingly, the manipulation of context itself. Frontier model prompt injection and reasoning-layer attacks exemplify this shift. The objective is no longer simply to penetrate systems, but to influence how systems interpret, prioritise, decide, and respond. These are exactly the types of attack vectors that benefit from agentic systems capable of learning, iterating, coordinating, and adapting at machine speed. They are likely to become some of the defining cyber battlegrounds of the next decade.

The pattern that emerges becomes clear. Attacks that rely on force become less attractive relative to those that rely on understanding. And that also changes the economics on both sides. Attackers are able to reduce the cost per attempt while increasing precision and speed of iteration, while defenders (i.e. organisations) reliant on older operating models remain constrained by governance, risk tolerance, and the need to maintain operational continuity.

One side is optimising for success probability, the other for acceptable risk. And the gap between those two positions is starting to widen. This is exactly why cyber is moving first into Agentic AI, because it is the one domain where those pressures are already acute, can easily overrun an organisation, and cannot be deferred.

In wrapping up, let’s look at these changes through the eyes of a typical customer where the architectural picture becomes even more tangible. Many organisations already have the core components of what could be considered an agentic security model today, even if they would not describe it in those terms.

Identity might be managed through platforms like Okta, access and policy enforcement might be handled through services like Zscaler, behavioural analysis performed by tools such as Exabeam, and the underlying network environment observed and instrumented by providers like Extreme Networks.

Individually, each of these platforms performs a well-defined function and delivers immense value in its own right. Kind of like the family of individual intelligence agencies in a national context. Collectively, they begin to look like a set of independent decision engines tasked with operating across the same environment. Each one is answering a version of the same question, just from a different perspective.

• Identity determines who someone is and whether they should be allowed to act.

• Access layers determine where they can go and under what conditions.

• Behavioural systems assess whether what is happening is normal or suspicious.

• Network layers provide visibility into what is actually occurring across the environment at any given moment.

What is often missing is a clear definition of how those decisions come together when they need to be made in real time. And outside of business-as-usual, when does that actually happen? Usually in the moments leading up to, or immediately following, a major event where timelines compress dramatically and systems, policies, and response mechanisms are forced to coordinate at machine speed.

This is where the role that IBM offers today becomes really really clear, because it is not about replacing any of those layers, nor is it about competing with them in a traditional sense. It is about sitting above them and orchestrating how those decisions are prioritised, resolved, and executed when the system is under pressure. This was the IBM vision.

Today, they are delivering on this vision using a combination of consulting, security software products, and the broader IBM watsonx AI platform. That is a very different proposition from traditional managed services, and it is one that most organisations have not yet fully come to terms with.

Today’s reality is that most organisations have assembled the components of an agentic decision environment without explicitly designing how that environment behaves. Decisions are being made across a number of platforms, but the logic that connects those decisions is often implicit, shaped by default configurations, integration patterns, or the operating assumptions of individual vendors. And then even if they are mapped, where do these decisions live? That is no longer sufficient in the agentic age.

It will continue to work for a while in environments where decisions are relatively slow and human oversight can fill the gaps, but it becomes increasingly fragile as decision-making accelerates and becomes more distributed. At that point, the problem is no longer one of capability, because the tools themselves are highly capable. The problem becomes one of control, and specifically whether the organisation understands what the system is allowed to do and how those permissions are exercised in practice.

What sits underneath all of this is not simply a shift in cyber, but a shift in the way enterprise systems operate. For a long time, the dominant model has been built around systems of record and systems of engagement, with the assumption that humans sit at the centre, making decisions and directing activity. Security has been no different. Yet as more of that decision-making becomes embedded within the system itself, the structure begins to change.

Identity starts to take on the characteristics of authority. Access becomes an enforcement mechanism rather than just a gateway. Behavioural analysis becomes a form of judgement rather than reporting. And network visibility becomes a form of awareness that informs real-time action.

Individually, these changes are subtle, but collectively they redefine how the system behaves. And once that happens, the question that sits at the centre of every organisation becomes much simpler and much harder to ignore. Who decides what happens next? IBM’s decision to retain and invest in managed security services now makes total sense in the context of their broader Hybrid, Data, AI strategy. It is the domain where the consequences of not being able to answer that question are immediate.

IBM appears to have recognised earlier than most that the future of cyber would ultimately become a problem of orchestration, governance, and machine-speed decision-making. In doing so, it may also have inadvertently provided the industry with its north star, and the clearest onramp for agentic adoption at scale.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

The thing is, when organisations ask questions about ROI and total cost of ownership, they aren’t just asking abstract questions about value. They are actually trying to understand what it costs them to run the business that way. And right now, no one has the answer. But more importantly, we can’t wait for vendors to solve this. We are unlikely to meet in the middle. So the work of deriving cost doesn’t disappear. It shifts to the customer. Which means rolling up our sleeves, making work visible, understanding how it actually gets done, and translating that into cost.

In many ways, this may reintroduce consultative services back into the centre of enterprise strategy. Because adopting AI at scale does not remove operational complexity. It has to expose it. So as organisations attempt to make execution measurable, governable, and economically visible, a new executive function increasingly appears to be emerging alongside it. Not simply to oversee AI, but to coordinate how work itself executes across the enterprise. What does that role become? What does it own? And could it ultimately define the path to sustainable AI ROI?

Share

Enterprise technology has long been framed as a transformation problem. Whether that starts with a conversation to review an existing program of work, or more analytically to get straight to upgrading systems or modernise into more flexible hybrid cloud environments better aligned to address the modern reality of disruptive business, it has always been an unwieldy problem. Now we’re being asked again to modernise the stack, and introduce AI.

And yet, despite unprecedented tactical investment in tech, business leaders are struggling to explain how their organisations are executing the real work of delivering products and services to customers in a better way today than just a few years ago. If anything, I’ve observed a decline.

I think what CEOs are actually facing is not a technology problem in isolation, but one of business cohesion and alignment. And they are looking to bring the divergent parts of their organisation back together into a single, coherent way of operating.

Yet at the operational level, many of those flows remain fragmented, inconsistent, and largely invisible outside the people directly involved. You can throw a rock in most organisations and hit a process held together by workarounds, tribal knowledge, disconnected systems, and manual intervention. The challenge emerging in the AI era is not simply automating systems, but understanding how work actually moves through the organisation in the first place.

That is why so many symptoms of our legacy operating models persist. For years, organisations have pursued stronger performance within individual functions, believing optimisation at the local level would naturally improve the whole. In reality, it has often increased the friction between functions, creating a breakdown in organisational coherence. Not just coherence around how work gets done, but around what the organisation is ultimately trying to achieve and how value is created across the enterprise.

If value itself were more visible, would the coherence problem be less severe? Would rising operational costs be easier to explain? There is a strong argument that the current race to apply AI is amplifying both problems by accelerating execution without a corresponding ability to measure how that execution translates into meaningful organisational value.

The challenge is that AI is accelerating execution faster than organisations can rethink the models they use to measure productivity, performance, and value creation. Yet the contours of the new model are beginning to emerge. Increasingly, enterprise economics are being shaped by four things. How work flows, how decisions are made, how often execution breaks, and how much of that execution should be automated. In that world, cost is no longer simply systems plus people. It becomes a function of flow, decisions, exceptions, coordination, and automation.

That’s the CEO constraint. A lack of visibility into how work actually executes end-to-end. Without that, measurement is partial, and management has no other method other than to become reactive. It’s not a new problem. Deming identified it decades ago. What is new is the extent to which modern business systems have evolved faster than our ability to meaningfully observe how work actually executes through them.

This is why the market narrative has shifted so aggressively toward new platform-based architectures. To manage this opaque execution layer across fragmented systems. But this introduces the second problem.

Technology can enable coherence, but it cannot own it. Any attempt to delegate that responsibility entirely to technology already has a name. Managed cognitive dependency.

Cognition is not coherence. AI can accelerate analysis, automate decisions, and increase execution capacity, but coherence requires alignment across operating models, governance structures, systems, funding priorities, and decision rights.

And increasingly, no single role inside the enterprise is truly accountable for that outcome. Years of decentralised technology buying, fragmented digital programs, and function-led transformation have distributed execution authority across the organisation while weakening ownership of enterprise-wide coherence itself.

The CIO still owns systems. The COO still owns operations as they are understood (though I would argue that operations have fundamentaly changed). The CFO still manages cost. The CDO where they exist, still own data. But no one owns how work actually flows across all of it. So while the CEO, and everyone else, feels the problem, the organisation cannot resolve it.

The problem is not merely a leadership gap. It is a structural shift in organisational execution itself. But structural fragmentation does not eliminate the need for leadership. It makes coherence, alignment, and accountability even more dependent on it in a world where the technology stack is now multi-platform, where TCO models have moved from systems to execution and where continuous disruption has become the norm. That combination creates a new requirement.

I’ve increasingly observed calls for a new kind of executive role, and they are beginning to emerge. Last week’s appointment by the Commonwealth Bank of Professor Mary-Anne Williams as its first Chief AI Scientist is one example.

It will be interesting to watch how these positions evolve because I suspect they are not ultimately about technology ownership alone. Nor are they simply transformation or delivery roles. What organisations increasingly appear to need is a single point of accountability for enterprise execution itself.

In many ways, I think better external framing for the role is as a Head of Execution. Someone accountable for how modern work flows across the organisation, how decisions are coordinated, what execution costs, where friction accumulates, and how increasingly AI-enabled operating models scale coherently across the enterprise. That’s no small task.

Whatever the final title becomes, this is fundamentally a role about operationalising execution as a measurable discipline. It will need to redefine the “unit of work” inside the bank, instrument execution end-to-end, and expose the economics of how work actually flows through the organisation. Not just systems costs or labour costs, but the true cost-to-execute across decisions, exceptions, orchestration, coordination, and automation.

In time, that likely means establishing standardised execution patterns across functions and translating execution performance into financial terms the enterprise can govern at scale.

At an operational level, this program of work cannot remain abstract for long. It will materially shape the bank’s technology choices over the coming decade. Because once an organisation begins instrumenting execution, a fairly clear maturity path emerges. First comes making work visible. Then establishing baseline cost, flow, and performance metrics. From there comes optimisation, orchestration, and ultimately measuring ROI as a function of how work actually executes across the enterprise.

It becomes easy to see why investments in platforms such as Pegasystems would increasingly fall within scope, not simply as workflow tools, but as part of the operational architecture through which execution is standardised, measured, and scaled. Those are the kind of decisions that move markets. Not only for the platform providers themselves, but for the system integrators, advisory firms, and ecosystem partners aligned around them.

The natural progression is that as visibility improves and execution costs become measurable, AI and automation can be applied more broadly, unlocking higher ROI and creating a compounding cycle of operational improvement. In effect, this role would do what no current executive role is fully designed to do. It would connect work. Once those relationships become visible, organisational dynamics begin to change. Productivity, cost-to-serve, operating performance, and ultimately coherence itself can be tied directly to how work actually flows.

Only at that point, the promise of modern platforms and AI becomes real. Not because the technology changed, but because the organisation finally became capable of absorbing it. That is the operational signal the Commonwealth Bank has just sent to the market.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

For much of the past two years, the enterprise software market has been trapped in a strange contradiction. Artificial Intelligence has driven one of the largest waves of technology investment in modern history, yet many software companies have simultaneously experienced valuation pressure as investors ask uncomfortable questions about the future role of traditional applications. The logic is straightforward enough. If Large Language Models increasingly become the interface to work, if agents begin orchestrating tasks across systems, and if workflow intelligence becomes abstracted above the application layer, then what happens to the value of the application itself? Across global software markets, that question is weighing heavily on investor sentiment. Yet TechnologyOne has largely escaped this broader correction.

At first glance, that seems counterintuitive. TechnologyOne operates in precisely the types of environments many investors would traditionally classify as vulnerable to disruption. That is, ERP and Corporate Services with a focus on government administration and higher education systems. Finance, payroll, procurement and regulatory workflows are exactly the kinds of operational software domains that Silicon Valley has spent the better part of a decade claiming are ripe for reinvention.

And yet the market continues to reward TechnologyOne with premium style valuations while much of the broader, and arguably more sophisticated, software market wrestles with existential AI questions. While the client narratives have long been conflicted, this can’t be dismissed as irrational exuberance, and in many respects, the market’s confidence is entirely understandable.

Share

TechnologyOne has now spent decades building one of the most operationally embedded software estates in Australia and New Zealand, while continuing to expand, albeit very slowly, into the United Kingdom. It has also managed to execute a disciplined (guardrailed) transition toward SaaS. And it has consistently expanded Annual Recurring Revenue. And it has delivered predictable operational performance across cycles where many software companies have struggled to maintain credibility. In a market increasingly obsessed with execution certainty, TechnologyOne has become almost synonymous with operational predictability. That is what matters enormously in the current AI environment.

Because while AI creates excitement, it has also created its own uncertainty economy. Investors are trying to determine which software companies become stronger as AI adoption accelerates, and which become progressively marginalised as intelligence shifts upward into orchestration layers and agentic operating models. But right now, the market clearly believes TechnologyOne belongs in the first category.

The reasoning is relatively straightforward. TechnologyOne’s customer base consists largely of organisations that move carefully, procure cautiously, and carry enormous operational and regulatory complexity. Local governments do not casually replace payroll systems. Universities do not rapidly replatform finance environments because a new AI startup emerges. Government entities do not embrace architectural experimentation at the pace of venture-backed technology firms.

These organisations value continuity, governance, operational integrity, auditability, and institutional trust. In many cases, they also operate with comparatively low enterprise architecture maturity and constrained technology execution capacity. That creates exceptionally high switching costs and extraordinarily sticky customer relationships. This is precisely why TechnologyOne’s position became powerful.

The company does not merely provide software. More like an operational foundation where its target work exist inside tightly integrated environments, and where the value of the whole becomes greater than the sum of the modules. AI, in the form they hhave come to market with, initially strengthens that proposition because AI performs best when operating against unified operational context. This is exactly why TechnologyOne’s AI messaging to the investor market has increasingly revolved around the concept of a single data foundation and integrated operational visibility, and it is a proposition its customer base, en masse, has shown remarkably little appetite to collectively challenge.

But there is another critical factor driving market confidence that should not be overlooked. TechnologyOne is not merely telling its AI story but actively attempting to monetise AI directly and immediately through its existing installed base. The significance of PLUS is not simply that it adds AI capabilities to elements of the platform. The significance is that it creates a near-term ARR expansion mechanism inside one of the most captive customer environments in the Australian software market.

The company is effectively pursuing a large-scale value extraction strategy from its existing customer base by layering AI-enabled capability across operational modules customers already depend on. Analysts are increasingly focused on what TechnologyOne has identified as a A$80–90 million pipeline alongside PLUS-related expansion opportunities. Importantly, this is not primarily dependent on winning large volumes of new logos. In fact, there has been little whitespace success for them. The strategy is heavily oriented toward expanding wallet share within customers already deeply embedded inside the TechnologyOne ecosystem. Ultimately that matters because it changes the risk profile of the AI story.

TechnologyOne does not currently need to convince the broader enterprise market to completely rethink its operating model. It largely needs to convince existing customers that AI-enhanced workflows, reporting, automation, and operational assistance create enough value to justify additional ARR uplift across finance, HR, payroll, procurement, and related functions. In today’s environment this is a fundamentally easier commercial motion than attempting to displace competitors through aggressive net-new AI-led transformation programs.

In many respects, TechnologyOne benefits from what could best be described as a controlled operational garden. Its customer base sits within relatively protected environments characterised by long procurement cycles, low disruption velocity, and strong preference for trusted incumbents. The financial market likes this because it creates highly predictable ARR expansion pathways while simultaneously reducing the immediate threat of AI-driven displacement. That combination is exceedingly rare.

For the more mature-minded tech-enabled organisation, this still leaves a more complicated strategic question. That is, not whether AI strengthens TechnologyOne’s current position. In the short to medium term, it almost certainly does. But what happens in the next phase of enterprise AI. Because the second phase of AI transformation may look very different from the first.

I am more than happy to concede that phase one can reward AI-enabled systems of record. Phase two and beyond will still be a game of systems of execution. And that distinction matters enormously.

The current AI cycle largely revolves around embedding intelligence inside existing workflows and applications. But the longer-term direction of enterprise AI increasingly points toward orchestration. Multiple agents from multiple vendors operating across finance, HR, procurement, governance, service delivery, analytics, and operational platforms, dynamically coordinating work across an increasingly hybrid enterprise architecture. In that environment, the strategic centre of gravity begins to shift.

Under that scenario, the danger for traditional ERP vendors in an agentic era is not replacement, but longer-term relegation. Systems of record will remain essential. Finance still matters. Payroll still matters. Governance still matters. Regulatory compliance still matters. But the platform that orchestrates work across the enterprise may ultimately hold the greater strategic influence.

This is where the deferred long-term architectural question around TechnologyOne begins doesn’t go away. Their historical strength has come from controlling highly integrated operational environments inside relatively bounded domains. But agentic orchestration increasingly rewards openness, extensibility, workflow abstraction, and cross-platform execution capability. These are fundamentally different strategic muscles.

Companies such as ServiceNow, Salesforce, Microsoft, and IBM are all increasingly positioning themselves around this higher orchestration layer. Not merely as application vendors, but as operational coordination platforms. These are strong lead indicators of future change. But that does not mean TechnologyOne suddenly becomes vulnerable.

In fact, what we are seeing is that many organisations in local government and higher education may be years away from fully embracing these broader agentic operating models. Operational discipline, governance, and trusted execution still matter enormously. The market may also be correctly recognising that the next five years matter more than the next fifteen.

TechnologyOne’s appointment of Barry Dietrich into the Executive Vice President of Sales and Marketing leadership role should materially strengthen the company’s ability to execute against its immediate AI monetisation opportunity through PLUS and broader installed-base expansion. In many respects, this makes it less a story about speculative AI disruption and more a trust-based story about disciplined value extraction from a highly embedded operational ecosystem.

TechnologyOne may ultimately prove that tightly integrated operational suites become more valuable in the AI era, not less. The market clearly believes that possibility today. But it remains without question, that the ground beneath enterprise software is shifting. And the unanswered strategic question is whether TechnologyOne is building the future operational orchestration layer of the enterprise, or strengthening the transactional foundations beneath somebody else’s.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

Forget who has the smartest model. I think the Agentic AI platform war must ultimately become an infrastructure benchmarking problem, not just a model quality problem. Businesses are really looking for who can prove the most efficient execution fabric for orchestrating work at scale. Based on what I saw this week at ServiceNow’s flagship event, Knowledge 26, we’re a long way towards that goal.

For context, consider the mainframe era. The brands that dominated did not do so because people initially loved or understood mainframes. Or for that matter could easily differentiate. They won because enterprises needed confidence that mission-critical transactional workloads could execute reliably, repeatedly, and economically. AI is yet to arrive at that moment.

To achieve that confidence, a key measure of differentiation for mainframes was MIPS. It was the measure of computational throughput for the transactional era that helped businesses see that the technology coupled with their implementation model of it, was competitive. There is not yet to be an equivalent for the agentic era.

But I think TAPS (Tokens Attributed Per Second) could be that measure. And unlike raw token counts and costs which are understood, let’s call them TRU (token resource units), TAPS would allow the measurement of orchestration, flow, sequencing, exceptions, and execution efficiency across a business process.

Read on if Servicenow’s hockey stick moment is of interest.

Share

I just spent another week in the US at another amazing AI event. And while it was compelling, from all the conversations I had with both execs and regular customer delegates on the conference floor, it was also more evidence that there is still something strangely incomplete about the current AI market.

Every week we hear another declaration that one model is now marginally better than another. No better example than while I was at Knowledge, Anthropic separately announced “dreaming” as a new feature. And then there are the ever more agent announcements, and more autonomy opportunitues, more OOTB workflows, and more partnerships. It’s like wading through mud.

The market is behaving as though intelligence and scale and feature adoption alone will determine success and adoption. But it won’t. Enterprises do not really buy intelligence. They buy execution.

That distinction matters more than most sellers currently seem to realise. Because buried underneath the noise of the current AI cycle is a much older pattern that has played out across every major transition in technology history.

The early phase of every infrastructure market is dominated by excitement, expansion and marketing narratives as vendors chase adoption velocity. Investors on the other hand chase enormous growth curves. Buyers are left with possibility, through conversations dominated by potential rather than operational reality. That is exactly where AI sits today.

That matters because pre-IPO and hyper-growth environments reward very different things than mature infrastructure markets do. ServiceNow CEO Bill McDermott calls this the AI blind spot.

Right now the market is rewarding model capability, developer adoption, ecosystem gravity, consumer awareness, token growth, and perceived inevitability. Operational efficiency is discussed but still secondary though it will ultimately matter more once the dust from the big $1T IPO bubbles hopefully settles this year.

In many ways this mirrors the early days of the mainframe era. Mainframes were not initially sold because enterprises deeply understood computational theory. Years ago I worked with a guy who escorted ANZ Bank executives on a multi-month delegation through the United States ahead of their very first major mainframe purchase.

The stories were wild. Endless vendor briefings, demonstrations, dinners, dude ranches, and political theatre. But underneath it all was a very serious question. Could this new technology platform reliably run the bank’s critical transactional workloads repeatedly, safely and at scale? The trip was ultimately successful not because the executives suddenly became computer scientists, but because they gained confidence that the operational heart of the bank could execute on this new infrastructure.

What followed was transformational. Banks needed payment processing. Then airlines needed reservation systems. Then Governments needed records management and retailers needed inventory control. The mainframe became the operational backbone of the modern enterprise because it created confidence in execution. But not all mainframes were created equal. So eventually that confidence became competitively measurable. Through MIPS. Millions of Instructions Per Second.

I also worked with one of Australia’s pre-eminent benchmarking specialists who was still generating millions of dollars a year for analyst firms well into the early 2000s doing benchmarking projects for the country’s largest telcos, banks and state governments.

Anyone that has ever seen a benchmarking engagement up close knows MIPS were never perfect. There were endless arguments about weighting, workloads, utilisation patterns and what was really being measured. But that was never really the point.

What MIPS gave the market was a comparative language for discussing throughput, scale and operational capability. A way for organisations to compare not just machines, but the transactional confidence underpinning their organisations. Importantly for this story, enterprises were not really buying mainframes. They were buying confidence that the business itself could execute better. That same transition is now emerging with agentic AI.

Today the market still talks about AI primarily as though it is software, largely because it is abstracted through modern PaaS architectures and wrapped in applications, copilots and conversational interfaces. But true end-to-end agentic systems are operational infrastructure.

And like every major infrastructure shift before them, their value will ultimately be judged against the thing they are seeking to replace or outperform. That’s not humans. It’s decision chains and transactional processes and more.

The real promise of agentic infrastructure is not that it thinks better than people. It is that it executes work across complex organisational systems more efficiently, consistently and adaptively than the operational models we have built during the previous technology era.

What organisations actually care about is whether work flows more effectively through their value chain which, at some point through decomposition, standardisation and transformation, ultimately arrives at workflows. And workflows, including yours dear reader, are ugly things ripe for disruption.

They are not neat diagrams in strategy decks. They are living operational compromises accumulated over years or decades. They contain approvals, escalations, retries, integrations, policies, exceptions, governance controls (hopefully), human intervention, shadow processes, duplicated decisions, historical baggage and institutional chaos layered on top of one another through successive generations of technology and management thinking.

That’s how we are arriving at a consensus that the hard problem is not generating intelligence through models. It is actually orchestrating the work those models support. Which is why I increasingly suspect the AI market eventually needs its own MIPS moment.

It needs to be something capable of measuring how efficiently tokenised work moves through complex, multi-functional orchestrated operational environments. Something like TAPS. Tokens Attributed Per Second. Not as a metric for models, but as a benchmarking input for workflows executing across agentic platform environments.

Because the real challenge is not isolated prompts or single-agent interactions. It is the coordination and orchestration of work flowing across functions, systems, humans, policies, APIs, approvals, exceptions and decision layers in real operational environments. TAPS would attempt to measure how effectively tokens are consumed, routed, contextualised, validated, transformed and completed as work moves through these interconnected execution chains.

Not just “how many tokens were used”, but how efficiently the operational fabric itself converts tokenised reasoning into completed outcomes across the enterprise. That’s the unproven point of differentiation and why the current market conversation still feels incomplete. And part of the reason financial analysts are gutting software stock (the blind spot).

Right now the industry is obsessing over token pricing, and model scale, and inference costs. But enterprises run workflows which create entirely different economic pressures.

What happens when agentic systems begin running finance, procurement, HR, customer operation, compliance, field services and government processes at scale? Suddenly the important questions become which platforms minimise retries, and which orchestration layers, regardless of architectural partnerships, reduce token waste. Or as was evident at Knowledge 26 this week, the current red hot moment is which AI platforms govern complexity and risk most effectively? That is not a model intelligence conversation anymore. It is one purely focused an operational throughput.

I thought Paul Fipps, Servicenow’s President Global Customer Operations had a good message. CIOs don’t want ungoverned custom software running around the enterprise. But it doesn’t stop there. CEOs expect that as a baseline, but they are actually looking for something more.

In the transactional era, measurable packets of data moved through enterprise systems. In the agentic era, packets of tokens move through orchestrated workflows. And each of these workflows involves a complex model of multi-vendor agent approvals, retrievals, escalations, classification events, retries and contextual handoffs

All of it becomes part of a living token economy where attribution and not unit cost will unlock the difference between good and great agentic infrastructure. The enterprise itself will slowly transform into a token-routing environment. And once that happens, measurement and benchmarking becomes inevitable. Not because analysts like me invent new acronyms but because procurement eventually demands operational comparability.

Boards and CFOs will not tolerate fuzzy economics forever once these systems become embedded into critical operational processes. They’ll want to know if this new tech can execute work more efficiently than the alternatives. That is when the market changes. That is when the hockey stick happens for ServiceNow. When they don’t just orchestrate and govern better than anyone, but when they unequivocally show they minimise operational drag across tokenised workflows.

In other words, the winners may end up looking less like AI companies and more like operating system companies. And I think that is something that suits a company like ServiceNow. That’s operational reality and totally aligned with the base that propelled them to #1 in category.

So perhaps we simply are not there yet. But I think we are very close. Perhaps the market first needs to pass through this expansion-era phase where narrative dominance matters more than operational efficiency. Perhaps we need to get through the growth cycle, the valuation cycle and the platform land grab before the harder industrial questions (and answers) arrive. Because history suggests they always do.

Eventually every infrastructure market matures. And when it does, the conversation shifts from what can this tech do to how efficiently can this technology run the world?

That was the real story behind mainframes. And I increasingly suspect it will become the real story behind agentic infrastructure platforms too. MIPS helped sell the transactional age. TAPS, or whatever its equivalent may be, may yet define the agentic one.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

Today’s AI narrative has been dominated by model breakthroughs, funding rounds, and product launches. It has felt like a glitzy, fast-moving, parallel universe. At times, more like watching the Oscars than understanding how work will actually change. Understanding, not shaped by Silicon Valley or Wall Street, has taken time to arrive.

I think what is starting to shift now is not the pace of innovation, but the quality of market thinking. The centre of gravity is moving away from the builders and toward the users who actually have to make this work.

From that perspective, whether you are an enterprise buyer or an institutional investor, these are ten observations about where I think the real AI market is heading, and why they will matter far more than the next model release or benchmark comparison.

1. The model is no longer the centre of the story

It initially looked like the model was the product. I think we are past that. Models are becoming powerful components. But they are increasingly interchangeable, increasingly abstracted, and increasingly hidden behind other layers. The centre of gravity is shifting upward into how models are applied, governed, and embedded into work, and the platforms that allow this to happen.

2. Software is still the distribution mechanism for AI

No matter how advanced the model becomes, the customer won’t just interact with a model. They’ll interact with software that relies on interfaces, workflows, approvals, forms, notifications, and dashboards. Bascially the same things that have always defined enterprise value. Nothing about that has changed yet. Because AI is an amplifier. It reinforce not replace the importance of the software systems through which work actually happens.

3. The “death of SaaS” is overstated, but the weak will disappear

We can debate the scale of the SaaS apocalypse, but surely all agree that there will be (and should be) casualties. A decade of proliferation has created a long tail of thin, single-purpose tools that exist largely because they could. Many of them will not survive. I’ve described this phase as “the application frat party” or “whatever-as-a-service.” Core systems, however, are not going away. If anything, systems of record become more important. They provide the structure, authority, and constraints that AI needs to operate safely. What disappears is not software itself, but the fragmentation. And that is long overdue.

4. The real enterprise problem is no longer integration. It is coordination

Connecting systems has been an observable challenge for most of my career, from the early days of “what the hell is middleware?” to today where APIs have largely solved that problem. But the harder challenge was never just about connecting systems. It was about following how work actually moves through a business. Defining who approves what, what triggers the next step, where decisions branch, how exceptions are handled, and how multiple functions interact in real time is still incredibly difficult. Where integration connects systems, coordination connects work, and delivers coherence back to the business. So to be valuable, AI has to sit inside a coordination layer.

5. Context matters more than data, and most organisations have lost it

At the exact moment AI arrives as a context-driven capability, most organisations have never had less clarity about themselves. Over the past decade, the rise of as-a-service has fragmented ownership, spread decision-making across functions, and steadily weakened enterprise architecture. Systems have multiplied, the CIO’s influence has diminished, and organisational coherence has declined as a result. At the same time, organisations have invested heavily in data lakes, warehouses, and pipelines, assuming that centralising data would unlock value. AI is now exposing the limits of that thinking. What remains is data without structure, systems without alignment, and organisations that struggle to explain how they actually operate end to end. That is the context gap, except it no longer a gap. It is a chasm.

6. Enterprise architecture is coming back, whether organisations are ready or not

The rise of SaaS decentralised technology decision-making (see #3). Buying groups emerged everywhere. Systems were acquired function by function. Architecture, as a discipline, quietly weakened. But AI does not tolerate that fragmentation. It forces organisations to confront how everything fits together and it exposes gaps, duplication, and incoherence. Architecture, that lost art, is no longer optional. It is being reintroduced by necessity.

7. The service layer is not shrinking. It is moving up the stack.

The belief that AI will shrink the services market and that automation will replace people, and the need for services will decline is just plain wrong. The service layer is actually becoming more valuable than the model layer. So the familiar pattern still holds. The one where the product arrives first, but the real value follows in services. The organisations that understand this will capture far more value than those focused purely on building models. Because AI doesn’t simplify the enterprise but it does make its complexity unavoidable. So the work moves, not disappears. It shifts toward far more demanding activities like stitching together models, workflows, governance, identity, and data into something that actually works inside an organisation. It is not implementation in the traditional sense but still services engineering.

8. Closed ecosystems will persist longer than expected

The case for open ecosystems built on a soft core and hardened shell has always been clear. But the market is not rational. Closed systems continue to win because many organisations are not equipped to manage multi-vendor complexity, so they default to simplicity. I see this consistently, particularly in government environments. That creates a paradox. Closed platforms are structurally weaker, but commercially durable. It is what keeps ERP vendors in the game. That durability buys time, but it is also finite. As the market evolves, remaining closed becomes harder than transitioning.

9. The AI bubble may need to dim before real progress accelerates

Right now, the model spotlight is too bright. It is drawing attention away from the harder, slower work required to make AI useful. That work is not in the models. It sits in process redesign, governance, data meaning, and organisational alignment. These are not headline topics, but they are the ones that determine outcomes. So the constraint is no longer capability, but focus. Progress will accelerate when the spotlight shifts, either because it fades, or because buyers (and investors) start dimming it themselves.

10. Most organisations are being sold the penthouse, not the front door

Vendors are pitching end-state visions replete with fully autonomous workflows, agentic enterprises and self-optimising systems. But organisations do not start there. They have always needed entry points in the form of clear, grounded, level one use cases that light the pathways into more mature adoption. Without that, the gap between ambition and execution remains way too wide (commercially and technically). The next phase of AI adoption will be defined by those who can show customers where to begin, not just where they could end up.

That’s it for now. So despite the tendency to believe that everything has changed (because this technology is so amazing), in reality, what is happening in the real economic markets, Monday to Sunday, is far more subtle. Almost recognisable?

What sits underneath these shifts is not a technology story at all, but a clear decision framework about where to invest, what to rationalise, how to architect, and how to execute. Things like “treat models as components, not strategy”, and “shift services investment up the value chain” and “rationalise the long tail of SaaS” are very practical steps to take.

AI extends what is possible, but it doesn’t rewrite the story. The fundamentals of software still apply, the complexity of organisations still matters, and the difficulty of execution hasn’t gone anywhere.

If anything, this may be the crowning achievement of the last 50 years of technology. It gives every organisation the potential to be great. But what happens next is not a technology question. It’s an organisational one.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

One of the more interesting dynamics in the enterprise AI and workflow market is how it is beginning to reshape established ecosystem relationships. Take, for example, that between IBM and ServiceNow. This is not a marginal partnership. It spans resale, implementation, co-selling, and increasingly joint positioning around AI and workflow. While no single value is disclosed, the economic weight is significant. In APAC alone, IBM’s influence across these channels is likely in the range of $200–300 million annually, with the global figure considerably higher. More importantly, the downstream value through transformation and services is materially larger.

With offerings like watsonx Orchestrate, IBM is increasingly able to challenge ServiceNow in sectors such as banking, telecommunications, and the public sector. Particularly in APAC, where complexity, legacy integration, and architectural control are critical. This introduces the potential for meaningful headwinds, possibly approaching $1 billion over five years at the upper end of scenarios.

I’m not saying this should be interpreted as a direct loss. But we should see it for what it is. An ongoing contest over who owns the execution layer. To be successful IBM does not need to displace ServiceNow wholesale. It only needs to win in environments where workflow is inseparable from data, infrastructure, and legacy control.

In APAC, where ServiceNow’s revenue base is approximately $1.5 billion and growing toward a $3–4 billion opportunity over the next five years, the battleground is not just Net New ARR. It is also NRR under pressure. Not just who owns the installed base (which is the old business model), but who captures the next layer of spend within it (the new business model). IBM’s architecture-led, distributed execution model creates a very credible techncial pathway to intercept new regional spend that might otherwise default to ServiceNow.

The likely outcome is a selective diversion of high-value growth segments, particularly in industries where architectural authority remains contested. Framed this way, a $300–700 million shift over five years could be a reasonable expectation, with a $1 billion scenario representing the upper bound if IBM consistently captures complex, AI-led transformation programs. Of course, that’s just head-to-head and doesn’t factor other major players in the AI and workflow ecosystem aggressively positioning for market share.

Regardless, the strategic implication is clear. This is less about platform competition and more about who defines how work is executed in the next generation of enterprise architecture. That is exactly IBM’s point of entry. Let me explain the breakdown.

I recently spent a few days in Bangalore at IBM’s Asia Pacific analyst event, Insights 2026. They apply transformation across a set of capabilities that, in reality, operate in multiple directions at once. So it included a broad sweep across AI, data, and hybrid cloud, alongside extensive conversations with consulting teams and enterprise clients, including captive centre operations customers.

These sessions are always valuable because you get close to the thinking and hear how the company wants to be understood. But they don’t present a fully resolved picture. They don’t explicitly show how the pieces fit together. That part is left open. And that creates a familiar challenge. The more comprehensive the briefing, the harder it becomes to see the underlying shape of the real strategy.

So while IBM didn’t frame it this way, sitting there it was already starting to become clear to me that what I was hearing wasn’t just about the convergence of their AI, Hybrid and Data portfolios. It pointed to something more structural. A convergence towards the same layer of the enterprise that platforms like ServiceNow have defined.

Then, on the flight home, somewhere between Singapore and Brisbane, that picture sharpened in an unexpected way. A targeted email campaign promoting discounted SME offers across IBM’s core portfolios.

At face value, it was unremarkable. Just low-cost entry points into watsonx, data platforms, and hybrid cloud. But in the context of the previous few days, it changed the question from “what does IBM sell?” to “what are those pieces designed to become when you put them together?” Because that’s no longer a product question. It’s a platform one.

And when you look at it that way, the answer becomes clearer. An SME would typically buy a packaged application from an ISV. IBM is now selling the components to build that application. So while I’ve been critical in the past of its liberal use of the word “platform”, in 2026 I saw clear evidence that it is starting to execute in a way the PaaS market, and its enterprise consumer base, will recognise.

At enterprise scale, IBM is not easy to read cleanly. The most useful signals don’t come from how it describes itself, but from how each layer of the portfolio is now being positioned. It spans too much of the market to fit neatly into a conventional category. In my own filing structure, it effectively sits under “Vendor” and then “Mega Vendor,” because anything more precise quickly breaks down. It is not just a cloud provider, not just a software company, and not just a consulting firm, even though all three remain central to how its capabilities are delivered and realised.

So rather than trying to categorise it, the better way to understand IBM is to look at how it is organising itself. Increasingly, that comes down to three core domains that the market typically treats as separate.

AI, represented through platforms like IBM watsonx, promise not just model access but governance, auditability, and control. Data, increasingly anchored in offerings like IBM watsonx.data, which attempt to unify structured and unstructured information into something that can actually support decision-making. And hybrid cloud, underpinned by Red Hat OpenShift, which provides the execution environment across on-premise and public cloud infrastructure.

Seen together, the portfolio has, for some time, felt like a collection of adjacent bets. Coherent at a strategic level, but more of a vision than an operating model. The “old” IBM, viewed through the lens of a large enterprise, typically required translation, alignment, and a level of architectural intent that only existed in pockets.

This is why IBM has often appeared complex and fragmented. Not because the pieces didn’t logically fit, but because, without the involvement of IBM Consulting, they were rarely brought together as a unified system.

That’s the enterprise view. But what happens when you strip all of that away? The SME lens forces a different question. What would I actually use here? And when you look at IBM through that lens, the portfolio stops behaving like a set of products and starts behaving like a system.

Because while an SME might ask, “what would I actually use?”, the reality is they are still dependent on everything that sits beneath it whether that is identity, data, workflow, integration, and governance. So the platform doesn’t go away. It just becomes invisible, and unavoidable.

Increasingly, those dependencies are being organised around how work is actually executed. From a distance, the inclusion of AI in these SME offers looks like a standard market move. Discounted access to models. Credits for experimentation. A pathway into generative AI. But through the same lens, the dependency becomes obvious.

In the IBM model, AI is not intended to sit on top of work. It is designed to operate where work is already structured, observable, and governed. And that is the catch. Because creating that environment still requires significant process and data alignment. And these are areas where IBM Consulting has been deeply involved. The platform may now be visible, but it still needs to be assembled.

Without that, the models have nothing meaningful to act on. The outputs become inconsistent, the risk becomes unmanageable, and the promise collapses into novelty. So what looks like an AI entry point is actually another platform signal. AI is not something you add. It is something that emerges once work is already structured, observable, and governed. In other words, it operates at the level where work is actually executed.

The same pattern holds when you look at the data layer. At enterprise scale, the language of data fabric and lakehouse architecture suggests consolidation and opportunity. It offers a pathway to unify data and unlock value through analytics and AI. But at SME scale, it feels very different.

It exposes fragmentation, inconsistent definitions, and multiple versions of the same truth. Systems that were never designed to speak to each other are now being asked to form a coherent picture.

In that scenario, what is being offered is not just a platform, but a confrontation with reality. Data, in this context, has little value because it exists. It becomes valuable only when it can support execution and when it can inform decisions in a way that is consistent, auditable, and repeatable.

And then there is hybrid cloud, through platforms like Red Hat OpenShift. Outside of enterprise, the ability to run workloads anywhere and move between environments is often framed as optionality. But even from an SME perspective, that framing eventually breaks down.

You don’t just need hybrid cloud when your environment is already complex. Or when you are operating across multiple systems, multiple vendors, and multiple constraints. Or when control becomes more important than simplicity. You also need it through outages, geopolitical disruption, or supply chain failures like we’ve seen in recent weeks across the middle east. In that sense, hybrid cloud is not about infrastructure flexibility. It is about maintaining control over how work is executed in the presence of complexity.

Across all three layers, the pattern is the same. The value only materialises when it connects to how work is actually carried out. But then when you bring these three layers together, something shifts. AI, data, and hybrid cloud stop behaving like separate portfolios and become interdependent. All of a sudden this is not a collection of software infrastructure tools. It is an operating model.

So what does this mean in practice for the enterprise buyers and architects currently redefining their AI and platform strategies? Because once you strip it back, the market is no longer organised by products. It is organised by where work begins, where it is executed, and who controls it.

Microsoft still owns the broadest entry point for now. It sits where people interact with work. It makes AI immediate, accessible, and embedded in the tools that organisations already use. That position is not under threat in the near term. If anything, it is expanding, albeit with some opaqueness around its new AI-enabled E# licensing models.

ServiceNow has spent the last decade building something different. Not productivity, but execution. And the last two years re-establishing around being the platform where work is defined, routed, governed, and completed. Its language is increasingly consistent across workflows, data models, and systems of action. More recently, promoting the idea of a control plane for how work moves through an organisation.

What became clear in Bangalore is that IBM is now speaking a language that is uncomfortably familiar. Its framing is no longer confined to infrastructure, data, or models, but is now led by workflows, orchestration, and control towers. It is moving into observability of the execution layer and into the domain of how work itself is coordinated and governed. This is not a coincidence. It is a signal.

Because it means the market is no longer cleanly segmented into layers. It is converging around the single question of who owns the system that defines, governs, and executes work? ServiceNow were first to get that right.

In that context, IBM is no longer a deep architectural layer. It is reaching upward, attempting to connect its strengths in AI, data, and hybrid cloud into a model that can participate directly in execution. At the same time, ServiceNow is moving downward by strengthening its data foundations, expanding its AI capabilities, and positioning itself as more than just a workflow engine. Microsoft continues to expand laterally, embedding AI and automation into every surface where work begins.

So how does it stack up? I think IBM’s integrated portfolio spanning AI assistants and agents, middleware, data services, hybrid cloud, infrastructure, and its broader ecosystem, brings something genuinely distinct to the agentic workforce discussion. You can also read some of my earlier thoughts on IBM here (Agentic AI’s Narrow Door).

Right now, its strength lies in environments where complexity, regulation, and scale demand control. Where governance is not optional, and where execution must be observable and auditable across fragmented systems. As the conversation shifts from data sovereignty to AI sovereignty, that positioning becomes increasingly important.

Even more so when you consider that, according to IBM’s own recent CEO study, fewer than 16% of organisations have deployed AI at an enterprise-wide level. But the direction of travel is clear. It is no longer enough to provide the layers beneath work. Rather, every major platform is now moving toward owning the system through which work is actually carried out.

If a further proof-point was needed, IBM’s rollout of the watsonx.data Context Layer, effectively a context graph and semantic ontology layer, is another clear signal of where it is heading. This is a great move by IBM and is not just about improving data access or analytics but about defining how data relates to work.

And that moves IBM directly into territory that has historically been owned by ServiceNow. Because at its core, ServiceNow’s strength has never just been workflow. It has been the ability to model the relationships between services, systems, and processes in a way that allows work to be consistently executed.

What IBM is building through its context layer is not just a data capability. It is an alternative path to the same outcome ServiceNow has been pursuing. ServiceNow started with workflows and built a data model (CSDM) around them. It is an area I’ve long thought they have underplayed and under-promoted so it will be interesting to see how hard and visbily IBM will push Context Layer. They seem to be starting from the opposite direction by establishing the relationships between data, systems, and entities, and using that to inform how work should be understood and executed.

It is a subtle shift, but an important one. Because while literally everyone is now talking about workflows, the platform battle is shifting from “who runs workflows” to “who defines what things mean.” So it clearly spotlights that whoever orchestrates work must be able to define the ontology of the enterprise itself. That is the layer that ultimately determines how work can be executed, governed, and scaled. I’ve previously written about that here.

IBM’s approach to identity within this model also becomes critical. Whether it aligns to an existing framework or establishes its own will determine whether it converges with platforms like ServiceNow, or creates a parallel model of enterprise control.

Returning to the original arc, for most small and mid-sized organisations reading that marketing email, IBM’s SME offers would feel like overreach. Too much capability. Too much implied complexity and too far removed from immediate needs. And in many cases, that instinct will be correct. That’s why IBM Consulting features so prominently in so many IBM enterprise deals. But that is not the point.

Because what appears excessive at small scale is necessary at medium to large scale. And more importantly, it reveals the direction of travel. We are moving away from a world where organisations buy software to support tasks. Towards a world where platforms define how work is structured, governed, and executed.

So at this inflection point, the easiest way to consider IBM’s relevance in the platform market is not to start with its largest customers, but to strip the problem back to its simplest form. Because when you do, what remains is not a story about AI adoption or cloud migration. We are well past that. It is a highly trusted and capable transformation blueprint for what work will look like in the next decade.

About Councilio

Councilio provides independent technology analysis, executive briefings, keynote speaking, workshops, and strategic advisory services.

Learn more: councilio.co

I’ve been asked a lot about TechnologyOne’s AI offering, PLUS, since it was announced in October last year. Most of the questions start in a similar place. How does it compare on price? What does it actually cost to run? And increasingly, how well does it fit, architecturally and commercially, against other AI platforms? They’re good questions. But the moment you start pulling on PLUS, you’re analysing something slightly deeper than an AI product. You’re analysing the business model underneath it.

It has taken a while, but everyone is comfortable calling TechnologyOne a SaaS company now. They offer a form of cloud delivery, they have built into a subscription revenue business model, and they provide continuous updates. On the surface, it fits. But there has also always been a quiet mismatch sitting underneath both the architecture and the economics, and AI will increasingly expose it.

TechnologyOne doesn’t price like SaaS in the way the market now understands it. It anchors pricing to the economic footprint of the organisation (rateable properties), not the consumption footprint (users) of the system. That has allowed it to avoid the classic SaaS mechanics of seat expansion, licence optimisation, and user churn. It behaves less like a SaaS platform and more like a utility tied to the size of the council. That has allowed it to be stable, predictable, and highly effective in the ERP era.

The infrastructure architecture reinforces this position. Running on Amazon Web Services with managed environments is not the same as operating a fully shared, multi-tenant runtime. Platforms like Salesforce and ServiceNow are designed to share everything except the data. That’s what allows them to align product, pricing, and scale so tightly. TechnologyOne sits somewhere different. A single code line, delivered as a service, but with enough separation in the environment to justify a different economic model.

None of this is a flaw. If anything it explains their success both in the customer and equity markets. In fact, it even explains something more important. It explains why they haven’t had to change. Local government is a market defined by stability, low risk tolerance, and slow structural change. TechnologyOne’s model has been almost perfectly tuned to that reality.

They’ve captured stable customers and delivered predictable growth by leveraging high switching costs. It has allowed them to modernise delivery in their own way and at their own pace, but without needing to modernise economics. They moved to cloud, adopted global SaaS language, and evolved the product without fundamentally disturbing how company (i.e. investor) value was captured. That has worked while the unit of value has remained stable. But now AI is changing the unit of scale.

In a traditional ERP environment, system activity is a rough proxy for organisational size. More properties equals more transactions equals more staff interactions. You could price against the size of the council and be broadly aligned to the amount of work being done. But in an AI-driven environment, that link breaks.

A council with 10,000 properties could generate vastly different levels of system activity depending on how aggressively it adopts automation. One might use AI to assist staff at the margins. Another might automate entire decision pathways, compliance processes, and customer interactions. The difference in system workload is no longer marginal. It’s exponential. That creates a structural tension inside the current model.

Therefore if pricing remains anchored to properties while the volume of work being executed by the system increases dramatically, the cost to serve rises without a corresponding increase in revenue. That’s why the cost question about AI is so important. AI can pretty quickly stop looking like a margin enhancer and start to look like a margin risk. Not because the technology doesn’t work, but because the commercial model isn’t designed to capture it. That is, unless those costs are pushed on to the client.

This is why the shift we are seeing from vendors like Microsoft and ServiceNow and Salesforce is so important. Their pricing models are moving, however imperfectly, toward consumption signals like per-user copilots, per-workflow execution, and usage-based constructs. They are aligning to the unit that actually scales in an AI world. That unit is “the work”.

Which naturally leads to the question many are now asking. Can TechnologyOne partner its way through this? On the surface, it’s a compelling path. Let Microsoft, ServiceNow and others provide the high-frequency AI execution layer. Let TechnologyOne remain the trusted system of record. Integrate the two and move forward without disrupting the core model. And in the short to medium term, that works.

In fact, it’s already happening. In many cases, it appears to be driven by the customer base themselves, rather than centrally led by TechnologyOne. Councils are experimenting with AI well beyond PLUS. Whether that be through Microsoft layers, or automation through ServiceNow, or marketing or customer engagement through Salesforce, or even content coherence through Glean. All of whicvh give broad access to every LLM available in the market, at the pace of the market.

Meanwhile TechnologyOne continues to defend its anchor position in finance, regulatory, assets and core records. This is creating a clean conceptual split between systems that store the truth and systems that act on it. But while that split holds for a long time if architected by the client, it doesn’t hold forever in a partner model.

Because the system that sits in the flow of work becomes the one handling requests, and decisions, and the automation, and the interactions and inevitably becomes the centre of gravity. It begins to own the experience and shape the workflows and captures the operational data that increasingly defines value. It becomes obvious to everyone where the value lies.

Over time, the commercial fear is that the system of record risks becoming a dependency rather than the platform. So partnerships, in that sense, don’t resolve the tension. They do defer it, and in other cases, amplify it.

Regardless, the volume of work still increases and the cost to serve still rises. But the value created by that work begins to accumulate in the partner layer. Which means the conversation about PLUS, and about AI more broadly, cannot stop at capability or partnership but has to come back to the operating model.

TechnologyOne will continue to introduce AI capabilities. And it follows that every successful AI capability will increase the amount of work being executed by the AI system. In the current model, that increased activity has no natural economic expression (because TechnologyOne’s model doesn’t price activity). That creates a disconnect. System usage can scale rapidly, while revenue remains anchored to relatively static measures like the rate base.

In the short term, we see what’s happening now. AI modules drive an uplift in ACV and ARR at contract renewal time. But without a mechanism to price the underlying activity, that uplift risks being episodic rather than structural.

Over time, this puts pressure on both Revenue Per Customer (RPC) and Net Revenue Retention (NRR). Revenue per customer becomes even less reflective of actual system usage (broadly tracked today as module access), and expansion depends more on discrete pricing events than on organic growth in workload determined by cost.

That leaves two options. They absorb the increasing cost to serve, or introduce new charges to recover it. Neither is comfortable when undertaken as “make good” or “reckoning” provisions in the contract. Especially where pricing predictability has been a core tenet of the sector. But also in a market where competitors are aligning pricing more closely to consumption, it’s definitely not a position that holds indefinitely.

Which means the question is no longer whether TechnologyOne can “do AI”, we’ve already seen it can at some level. The question is whether its current model can absorb it. And that’s where the idea of “reinvention” starts to appear.

But this isn’t reinvention in the way people instinctively think about it. There will be no collapse of the current platform or a sudden pivot away from ERP. It can be something far more subtle, albeit more difficult. And that is realignment from how value is captured to how value is now created.

Because for most of its history, TechnologyOne has been able to anchor value in what a council is based on its size, structure and footprint. But AI shifts that anchor point. Value will increasingly sit in what the system does and the decisions it makes and the work it executes and the interactions it automates.

That shift doesn’t require TechnologyOne to stop being what it is. But it does require it to finally evolve and improve how it participates in that work. It could be a commercial model that introduces a second axis tied to usage or automation. Or a product narrative that moves from system of record to system of execution. And over time, an architectural direction that supports better higher-frequency, lower-friction work across the platform. So also improvements in the underlying content, service and identity ontologies.

Partnerships can help accelerate that journey by bringing capability and keeping TechnologyOne connected to the flow of work. But they are only a bridge, because ultimately, this is not a question of capability but one of realigning from pricing based on what a council is to pricing based on what a council’s systems actually do. That is why property-based pricing, which once looked like a perfect fit for the sector, now starts to look less like a strength, and more like a constraint on its future.

The important thing for councils to understand is that vendors will evolve because they have to. But councils don’t need to wait for that evolution to be complete. The ones that move early won’t just adopt AI faster, they’ll define where value sits in their environment before someone else does. And in relation to AI, that may be the most important decision of all.

Agentic AI works today. But for most organisations it remains in a narrow band of work. The net result is that the market is misreading the signal. Because it is focused on the destination, the level 5 or 6 maturity benchmark, and not the entry point or the path to get there. Enter Salesforce’s SMB play.

Agentic AI looks like it’s working, especially in smaller organisations. It is quick to deploy. It is easy to demonstrate. And it can deliver immediate gains in productivity. It also fits neatly into some workflows where speed matters more than consequence. It’s the kind of model I wrote about in TechnologyOne’s PLUS Moment.

And if you read that as the starting point, not the end state, the current wave of success tells you something important. It does not tell you that agentic AI is solved. But it does tell you we’re seeing the earliest stage of a much longer progression. One that will take years to unfold as organisations move from low-consequence efficiency into work that carries real consequence and accountability.

I feel like this is the part of the story the market continues to gloss over. Not just because we’re misreading the results, but because of how the story is being told. Vendors are conditioned to sell the future. The fully autonomous state. End-to-end orchestration. Agents operating confidently across the business. That’s what I refered to as Level 5 and 6 maturity in an article last year called Have You Tried Our New AI Agent?

And it’s not just a marketing choice. It’s a market expectation. Boards want market share, and category leadership and to know they are going to win. So the narrative gets pushed forward where the destination is clear and the vision is compelling. But it sits orders of magnitude ahead of where most organisations actually are. And in doing so, the middle gets lost.

We jump from aspiration to demonstration, without spending enough time on what meaningful entry points look like. Where do you actually start? What kinds of work can be trusted first?

That’s where the confusion creeps in. The harder the market pushes the future, the less clear the starting point becomes.

Early success is too often interpreted as capability (for both the customer and AI solution provider), when in reality we’re just seeing what happens when you apply autonomy to low-authority workflows. It’s not that every AI technology being pitched is mature. It’s that the client work isn’t demanding enough to expose where it fails (at scale or higher levels of workflow complexity). Now, after a few years of headline-grabbing announcements, it feels like things are starting to settle down in 2026.

In a recent briefing with Salesforce, what stood out was their emphasis on strong AI adoption in the CRM SMB segment. Even allowing for some variation between Salesforce and ABS definitions, that’s a TAM of over 2.5 million businesses in Australia.

The smaller deal sizes in this segment make it easy to dismiss, especially for enterprise-focused sales teams. But that misses the point. If agentic capabilities are landing quickly and sticking, that’s a signal, not something to write off.

What if this wasn’t just the easy version of the problem, but the honest one? What if it shows where agentic AI actually works today, and not where a keynote says it should.

In smaller organisations, most workflows sit in a low-authority band. Fewer systems. Lower consequence. Lighter governance, often implicit rather than defined. That’s the space where agents work easily. They can summarise, draft, trigger, and respond. They can take small actions without needing a deeply structured model of the organisation behind them (see Whatever-as-a-Service). Identity exists, but it doesn’t need to carry much meaning. It’s present, but it’s very thin. In that technical environment, agentic AI feels easy. But what if we stop seeing this as a weakness and start seeing it as a starting point?

With its latest SMB offerings, Salesforce is demonstrating through commercial expression, something much of the market is missing. Agentic AI is not something you install. It is something you grow into. It is a maturity curve. And one of the most important of this time. Yet vendors continue to sell the destination. Yes, it’s technically valid and an exciting place to be, but also practically out of reach for most organisations today. Largely because we are in a transitional age of computing, and once again in too much of a rush to get there.

What’s missing is simply clearer starting points. And I can’t for the life of me understand why more vendors don’t take a beat and realign around this fundamental assumption. It’s time to rebalance the rhetoric. Less about where the plane or train is going and more clarity on the airport or platform where I can step on.

Share

When it comes to AI, it’s more useful to think of SMB not as a cut-down version of the enterprise, but as the starting point of one. That matters even more today, as enterprise buying centres have become increasingly fragmented. That is why the AI platforms that will define the next operating model are the ones built to carry organisations through that journey, as their work gains complexity and consequence.

Take identity as an example. A business might begin with a simple login, minimal governance, and little or no integration. At that stage, identity doesn’t meaningfully constrain agentic capability. Because both are naturally limited to low-consequence workflows. So you can adopt a solution that will allow you to perform Level 1 or 2 AI.

Now as the organisation grows, so does the complexity of its work. More systems are added (swivel chair work begins). More customers. More transactions. More exposure. Identity starts to evolve. From simple login, to managed access, to SSO, to a fully governed identity platform. Workflow becomes more structured. Decisions begin to carry a lot more consequence. That requires AI at a Level 3 or 4 maturity, perhaps now in the form of an Agent, to remain useful, and evolve with it. Ideally without changing the underlying AI platform.

That is a pathway. From point of entry to escalating maturity.

And I think this is the quiet logic behind Salesforce’s tiered SMB model. From free to starter, to professional, to enterprise, to more advanced agentic capability, the platform and its licensing model, is not just scaling features. It is scaling the step conditions, up to a tipping point, from which AI in the form of agents, can be trusted to execute.

For big or small organisations, that journey can take years and reflects something deeper about Agentic AI. That it is not just about what the agent can do but about what the organisation is ready to let it do.

This is where the broader market narrative starts to quickly break down. Because we’ve become conditioned to look for immediate proof of value of high-maturity capabilities. Where only fast adoption and instant capability can deliver visible impact within a reporting cycle.

But that’s not how platforms like Salesforce create value. They create it over time. As organisations grow into them. As their work becomes more structured. As identity matures. As governance strengthens. As more consequential workflows move onto the platform. The value compounds. Not because the technology changes overnight, but because the conditions for using it evolve.

That kind of progression doesn’t show up cleanly in quarterly narratives. But it is exactly the kind of pattern long-term investors have always looked for. The ability to build with the customer, not just sell to them. In that sense, the slower journey is not a delay. It has always been the model.

In the early stages of that journey, the scope of what can be trusted remains narrow. Which is why most of what we are seeing today is still operating in the efficiency layer. AI can remove some friction and speed up some navigation. Maybe even reduce the effort required to move between systems. But it is still operating around the edges of work.

This doesn’t change the view that the real AI shift begins when agents move into workflows that carry consequence. At this level of maturity (5 or 6) a payment is no longer just a transaction, but financial authority. A service request is no longer just a ticket, but a commitment tied to service levels, risk, and compliance. And a system change is no longer just an update, but a controlled alteration to a live operating environment (managed behind the scenes by a service ontology).

Now the agent is no longer assisting. It is participating. And participation requires structure. This is where identity changes form. It stops being about access and becomes about authority. Who can act, under what conditions, with what level of accountability. Without that, the agent can suggest but it cannot execute. This is the boundary most organisations, nor technology vendors, especially legacy ERP vendors, have not yet crossed.

Not because they lack technology, but because they have never fully defined how work actually happens. It sits fragmented across systems, buried in process documents, or held in the heads of individuals. More mature Agentic AI exposes that immediately. It doesn’t fail because the model is weak but because there is nothing coherent for it to act within.

But nobody starts there. And this is why the SMB success stories matter right now. Not because they prove agentic AI is simple. But because they show where it starts. The real test is for both the vendor and the client is still going to be what happens as those organisations grow. Because that is where agentic AI stops being easy. And starts becoming accountable.

So the winners will not be the platforms with the simplest generative interface or the most capable agents. They will be the ones that understand where to start. The ones that can meet organisations at their point of entry and grow with them over time. The ones closest to the flow of work. The ones that can bind workflow, identity, and data into something that can actually hold responsibility, without requiring constant reinvention (or system replacement) as that work matures.

Two comments came up in an analyst call this week that immediately stuck with me.

“Customers want a neutral, independent identity platform
that can be easily deployed and support the maximum number of use cases,
particularly as they begin to introduce AI into workflows and business processes.”

“Agents are emerging as a new identity type,
and organisations now need to figure out how to govern them.”

It took me a few minutes to work out why they didn’t quite land with me. When taken in isolation both are true. But I think they also make it too easy to assume the problem begins with identity. And that’s the part worth challenging upfront, for the benefit of both customer organisations and technology providers.

The first could lead someone to assume that organisations are ready to drop AI into workflows, as if those workflows already exist in a coherent, structured form. They don’t. The second could lead someone to assume that agents are primarily a governance problem, as if their existence is what creates the need for control. They are not. In reality, I think both are downstream effects of something else entirely. Work.

Organisations do need a neutral identity platform. But not because they’ve decided identity is important. They need it because work has become fragmented across systems, channels, and now increasingly across humans and machines. The moment work starts to span boundaries, identity becomes operationally difficult. The same applies to agents.

Agents are not interesting because they are a new identity type. They are interesting because they are a new form of work execution. They take actions. They trigger processes. They make decisions, or at least simulate them. That is what forces identity into the conversation. Not their existence, but their behaviour.

If an agent never acted, it wouldn’t need an identity. It would just be software. The identity requirement emerges the moment the agent participates in workflow. This is the context that often goes missing. No work, no identity.

The identity platform market is not expanding again because identity is evolving. It is expanding because the nature of work is changing. The last time identity surged was with SaaS, when the problem was access. It was all about getting the right people into the right systems.

This is very different. AI is not increasing access. It is increasing action. More interactions, more autonomy, more delegation. Identity is being reintroduced as a problem because something now needs to govern all of that movement.

That’s why the idea of a neutral identity platform resonates and why platforms like Okta are a strong day one choice. Not because neutrality is inherently valuable, but because organisations are struggling to maintain control across an increasingly heterogeneous environment. Multiple systems, multiple vendors, multiple execution layers. Identity becomes the only thing that can sit across all of it.

But even then, it is still responding to something else. The deeper question is not how to govern agents. It is what work those agents are being asked to perform, and whether that work is actually understood, structured, and worth automating in the first place.

Because if the underlying workflows are still fragmented, manual, or poorly defined, then introducing agents does not create value. It simply accelerates inconsistency. And wrapping that in a strong identity model does not fix it. It just makes it governable. Yo end up with controlled chaos.

This is the pattern that is starting to emerge. Identity vendors are correctly identifying that the world is becoming more complex, more distributed, and more autonomous. They are positioning identity as the layer that brings order to that complexity. But the complexity itself is not coming from identity. It is coming from work that has outgrown the structures designed to support it. Which leads to a more grounded way of thinking about it.

The next phase of the market will not be defined by identity alone, but by how clearly it is connected to the work it is there to govern. The identity vendors that recognise this best, and are most articulate about anchoring their story in workflow, action, and outcomes, will find themselves riding a much larger wave. Because identity does not create the demand. Work does.

Identity scales as work scales. It becomes more critical as work becomes more distributed, more automated, and more autonomous. But it is always responding to something else. So the opportunity for identity providers is not to compete for primacy, but to align more directly with where value is actually created. To move closer to the systems and platforms where work is defined, executed, and increasingly delegated to machines. Done well, this will not diminish identity but amplify it.

Because when identity is tied to action (who or what is allowed to do something, not just access something) it becomes part of the operating model, not just the control layer around it. And that is where the market expands. Agents make this unavoidable. They are not a new identity problem, but a new form of work execution. And as that work scales, identity is pulled into it. Which is why identity and workflow cannot stand apart for long.

An identity layer without coherent workflows has nothing meaningful to control. A workflow layer without identity has no safe way to scale. They are not competing abstractions. They are interdependent as I called out in the Whatever-as-a-Service post.

The organisations that recognise this will stop treating identity and workflow as separate domains, and start treating them as two sides of the same operating model. One defines what can happen. The other defines who or what is allowed to make it happen. This is where the market is heading.

Platforms like ServiceNow and Salesforce are expanding identity capabilities because they need control over the work they orchestrate. Identity providers like Okta are moving closer to workflow because they need context to remain relevant. The centre of gravity is shifting toward the intersection. And in an agentic world, that intersection becomes the system.

Agents don’t just need identities. They need roles, permissions, and boundaries tied directly to the work they perform. That requires more than authentication. It requires a shared understanding of entities, states, and actions. That’s an operational ontology that both identity and workflow systems can interpret in the same way.

Without that, identity and workflow drift apart. With it, they reinforce each other. Which is why the likely winners will not be those who dominate one layer in isolation, but those who align both. Not simply around access or automation but around coordinated, governed action. That is where the real value sits.

Behind all of this sits a relatively well-defined group of control-plane identity providers. But the strongest will no longer be those with the best integrations. They’ll be the ones closest to the flow of work. For paid subscribers, I’ve put together a short view of where I think the major vendors sit today against this shift.

Read more

Everyone says you can’t do artificial intelligence with bad data. And at face value, the argument seems reasonable. But what is the answer then? We can’t simply keep repeating that transformation will take longer than expected without a clearer explanation of how fragmented enterprise data, with all its inconsistency and incompleteness, is supposed to be fixed before AI can safely operate.

And what about the organisations with messy data that are already beginning to see real operational value from AI platforms? Not through chatbots or document summarisation, but through automation, orchestration and decision support embedded directly into enterprise workflows. How does that make any sense?

Well, as it turns out, it is precisely this contradiction that reveals the deeper truth. The biggest barrier to AI inside most organisations has never been bad data. It is that they have never clearly described how the relationships actually operate between people, systems, approvals and the work that flows between them.

Enterprise technology has long been dominated by the simple assumption that systems require clean data. This belief emerged from the era of enterprise resource planning where systems were designed to execute deterministic transactions. If the underlying data was inconsistent, the results could not be trusted. That was fine for a generation, but the same assumption has carried forward into the AI era almost without question.

But artificial intelligence does not simply operate on tables and records. It operates on meaning. And that requires us to stop thinking about this technology in the way we have managed enterprise systems for the past thirty years.

For AI to perform real work inside an organisation, it must understand several things at the same time. It must know what objects exist in the enterprise, how those objects relate to one another, who has the authority to act on them, and how work moves between them. That’s what I mean by understanding the relationship structures of the organisation itself.

We’ve all worked for organisations that have never formally defined these relationships. They exist implicitly inside systems, processes and job descriptions, but rarely as a coherent model of how the enterprise actually functions or behaves. That model has a name. An ontology. And without that model, artificial intelligence is forced to improvise.

Ontology is a word that sounds abstract and so it is rarely used even by the companies that have great ontological solutions, and almost never in executive company. But the concept is straightforward. It is simply the structured representation of the things that exist in an organisation and the relationships between them.

A council, for example, operates around concepts such as citizens, properties, permits, assets and service requests. But the important thing is not simply that these records exist in systems (as a data model). It is how they relate to the work of the organisation.

A citizen may own a property. That property may require a permit before certain activities can occur. A permit must be assessed and approved by an officer. An inspection may later confirm that the conditions of that permit have been met.

Taken together, these relationships describe how the organisation actually operates, how decisions are made, who is responsible and how work flows across the council. Without that structure, artificial intelligence is left guessing or hallucinating. It can see the records, but it cannot understand the work.

Language models don’t fix this. Even connected to the right documents it can only summarise policies or generate reports. But what it cannot do alone is safely execute work across systems, departments and authorities. It does not know which team owns a service, which role must approve an action or how a process should unfold. Those rules do not live in the data alone. They live in the meaning of the enterprise. And meaning is the currency of AI. Said another way, AI runs on infrastructure but it operates on meaning. It’s just that most organisations have never actually modelled what their organisations mean.

So how did we get here? We can all agree that the cloud revolution dramatically expanded the number of systems inside organisations. It has been an appslosion! Sales teams adopted customer platforms, HR teams adopted workforce systems, finance departments moved to cloud ERP solutions and collaboration tools multiplied across the enterprise.

It’s a phenomenon I call “Whatever-as-a-Service”. A decades long application frat party. Connectivity improved dramatically but organisational meaning did not. It actually got worse. Each system introduced its own definition of the world. A “customer” in one application became an “account” in another and a “ratepayer” somewhere else. Integration platforms allowed data to move between these systems, but they rarely reconciled their underlying semantics.

We humans have quietly resolved these inconsistencies through experience and context, not architecture. But Artificial intelligence cannot. For AI to operate safely, the enterprise must first be represented somewhere as a coherent model of objects, relationships, authority and process. But where? This is where platforms like ServiceNow begin to change the conversation.

They have gradually assembled a structure that resembles an operating environment for enterprise work. Through frameworks such as the Common Service Data Model (CSDM) and the Configuration Management Database (CMDB), it is the role of the platform to model relationships between services, applications, infrastructure and teams. Incidents impact services. Services support business capabilities. Teams own responsibilities. Workflows define how requests are approved, escalated and resolved. What emerges is not simply a data platform, but a representation of how the organisation behaves.

You can see this clearly in some of the publicly available architecture diagrams. The ontological elements I am talking about in this article do not sit purely within a data model or the organisation’s systems of record.

Instead they appear in what ServiceNow describes as the semantic and intelligence layers of the platform. This is the connective tissue that ties enterprise data, workflows and AI together. It’s the secret sauce your systems architecture is probably missing if you want to do AI well.

That ontological representation of the business provides something more valuable than perfectly structured data. It provides operational coherence. And when relationships between services, teams and responsibilities are clearly defined, work can move predictably through the enterprise even if the underlying data describing those objects remains imperfect.

Incidents can still route to the responsible teams. Approvals can still reach authorised managers. Changes can still propagate through dependency structures. The ontology stabilises the environment. Which brings us back to the point. Perfect data may be desirable but ontology is essential.

Another critical component of enterprise ontology lies in identity systems. Platforms such as Microsoft Entra or Okta maintain a living map of organisational authority. They define who belongs to which teams, which systems individuals can access and which roles they perform. Those relationships quietly encode part of the enterprise’s operating structure.

For artificial intelligence, this authority layer is essential. An AI system cannot safely approve a payment, provision an employee account or modify a record unless it understands who is authorised to perform those actions. Governance therefore becomes structural rather than procedural. Security, identity and workflow together form the guardrails within which intelligent systems can operate.

Newer AI-native platforms are beginning to recognise this pattern as well. Systems such as Glean (in Australia, via Journ3y), construct knowledge graphs that map relationships between people, documents and projects across collaboration tools. Rather than attempting to impose a new security model, they inherit the permissions already defined in identity and collaboration platforms. If a document is accessible in its original system, it remains accessible in the AI environment. The insight is simple but powerful and this kind of leaning into the identity ontology approach underpins their growing success.

What this also reveals is that the meaning of the organisation can already exist in some places. It is simply scattered across systems, permissions and workflows. The challenge is therefore to surface those relationships. Once they are visible, the AI models themselves that feed the ontological structure become interchangeable. That’s why every big vendor is partnering with every AI model. It’s because ultimately the ontology becomes more important than the model.

The enterprise technology industry spent decades believing that clean data was the prerequisite for effective systems. It’s ok to say, structure the work first, and the data will eventually follow.

This shift carries another implication that the enterprise software industry has begun to recognise. Best-of-breed approaches to software work because humans reconcile the differences. AI cannot without a platform. Artificial intelligence operates far more effectively within environments where objects, relationships and authority are clearly defined. That environment increasingly emerges around platforms capable of modelling enterprise meaning. It’s part of the (very nuanced) argument discussing the “death of SaaS”.

Once such a platform exists, other systems begin to orbit around it. Workflows attach to it. Identity integrates with it. AI agents execute within it. Because ontology has gravity. Not AI. But AI is what captures attention. Ontology is what actually holds the system together.

This is why ServiceNow, Salesforce and Palantir are becoming strategically very very important. They are not simply applications. They are evolving to systems of meaning. And once an organisation begins to depend on a system of meaning, endlessly assembling best-of-breed tools (whatever-as-a-service) starts to lose its economic, financial, and technical appeal.

The bottom line is that we need to think differently about this. For decades we have pursued systems of record where the challenge was believed to be storing information correctly. Artificial intelligence forces a different question. It is no longer about where the data lives but whether the organisation has ever explained what the data means.

Therefore before machines can run the enterprise, someone must describe how the enterprise behaves. That description is not just a bunch of data schemas and lakes. It is an ontology. And the organisations that recognise this will discover that the biggest obstacle to AI was never bad data but understanding themselves.

That said, the answer to all this is not “buy a platform”, though I am glad that platform is now a broadly accepted conversation. It is more like “choose your entry point to platform.” That means identifying coordination failures within the business (problem statement). These coordination failures are usually the result of missing ontology.

I’ve included a simple example below. It reflects the kind of issue that commonly emerges during staff workshops and consultation when organisations are considering new systems. In one organisation we identified around thirty of these coordination problems in just a couple of days.

Let’s consider a common coordination failure like employee onboarding. In many organisations, provisioning a new employee across HR, payroll, finance, identity and IT service systems can take several days.

1. Coordination problem: Employee provisioning takes five days.

2. Desired outcome: Reduce onboarding time to two hours.

3. Why AI alone fails: AI cannot resolve cross-system discrepancies or coordinate approvals across multiple systems without understanding how those systems relate to each other.

The starting point is not the (new system or) platform. It is defining the ontology of the employee domain. That means identifying the core objects and relationships involved like employee, department, role, system access and approvals, and how they relate to one another. An employee belongs to a department. A department grants certain roles. A role determines which systems the employee should access and which manager must approve that access.

Once that structure is defined, the systems can be mapped against it. HR, payroll, finance, identity and IT service management platforms each manage part of that model, but none of them represent it completely on their own. The ontology provides the common structure that allows those systems to coordinate.

This is where platforms like ServiceNow become important.

Inside ServiceNow, those objects can be represented explicitly. An employee record may originate in the HR system, but ServiceNow can model the relationships between the employee, their department, their manager and the services they require. The platform’s data structures, including the configuration management database and related service models, allow those relationships to be expressed consistently across workflows.

Once the objects and relationships are visible, the missing capability becomes clear. Orchestrate the workflow across systems. That is the “simple” answer most execs want.

Using workflow automation and integration capabilities, ServiceNow can trigger provisioning steps across the relevant systems. HR creates the employee record. Identity systems provision accounts and access. IT service workflows allocate devices and system permissions. Finance systems configure payroll records. Approvals are routed automatically to the correct managers based on the defined relationships. So instead of integrating systems in isolation, the platform orchestrates the workflow across them.

At that point AI becomes useful. An AI agent can monitor the onboarding process, detect failures when provisioning steps break, investigate discrepancies between systems and recommend or trigger corrective actions. What previously required manual coordination across several teams can collapse into hours.

The larger point is that employee onboarding is only one example. Once organisations begin mapping coordination failures in the employee domain, a much broader pattern usually emerges.

In workshops and operational reviews across several organisations I’ve worked with, we routinely identify dozens of workflows that suffer from the same underlying problem. Work spans multiple systems, roles and approvals, but no single structure exists to coordinate them.

Examples include the creation and approval of position descriptions, recruitment workflows covering advertising, interviews, references and medical checks, employee induction processes including compliance training, leave request approvals, employee issue or case management, professional development requests and reimbursements, performance appraisal cycles, talent and skills searches, promotions and transfers, employee offboarding, and routine activities such as timesheet submission and approval.

Each of these processes touches multiple systems and involves several roles across the organisation, especially after more than a decade of whatever-as-a-service. Viewed individually they appear to be isolated workflow problems. Viewed together they reveal something more fundamental. The organisation lacks a shared model of how employee-related work flows across its systems.

This is where ontology begins to take shape. Each coordination problem exposes another part of the enterprise model that needs to be defined from the objects involved, the relationships between them, the authority required to act and the workflows that move work forward.

Over time those fragments accumulate into a coherent ontology of the employee domain. But that only works if the organisation has a platform capable of orchestrating across systems once those relationships are defined. Without that capability, the ontology remains theoretical. With it, the coordination problems begin to collapse.

That’s the power of a platform like ServiceNow. With or without clean data.

Australia’s major banks did not voluntarily embark on large-scale customer identity remediation. They were pushed. Regulatory tightening around KYC and AML reframed customer data quality from an operational concern into a compliance risk. In doing so, it exposed how much institutional confidence had been placed in identity records that were never truly verified. The banks aren’t now just fixing UX. They are protecting their licence to operate. Think about that for a second.

While the public narrative centres on financial crime, terrorism financing, and systemic integrity, something more structural is occurring beneath the surface. Compliance is quietly converting identity data into hardened economic infrastructure. Data that survives sustained regulatory scrutiny becomes economically different. It can be relied upon and automated against and shared internally without qualification. And it reduces the cost and risk of every downstream decision that depends on it.

And today that matters beyond compliance. Verified identity is not just a regulatory obligation. It is a prerequisite for automation at scale. Without trustworthy records, delegated work collapses back into manual oversight. But with them, institutions can safely transition toward machine-assisted and agentic operating models.

In a recent discussion with one of the Big Four banks, I was told they are actively validating more than 1.3 million unique customer records. Not enriching them. Not optimising them. Simply validating. Addresses. Employment status. Proof of identity. Duplicate CRNs. Parallel identifiers created across legacy systems. Records that were considered “good enough” for years, but are no longer defensible.

These are institutions with deep technology budgets, mature analytics capability, and decades of digital transformation behind them. They have invested heavily. They have modernised repeatedly. Yet even here, identity foundations required remediation at scale.

So, if that is the position of Australia’s most heavily regulated, best-funded data environments, what does that imply for everyone else? Well, a great deal.

Let me make this far less abstract using my own personal experience.

Over several months my bank sent repeated requests for updated KYC information (primed by their own regulatory deadlines). I delayed. I was busy, and yes, slightly stubborn. But I was also hesitant because I knew the friction that was coming. I couldn’t remember my CRN. Password? Yes. PIN? Yes. Account numbers? Naturally. I’ve been a customer for decades. But the CRN, that single identifier, isn’t something someone needs on a day to day basis.

I tried and failed to recover it online. The website process instructed me to '“Please call this number.” Urgh. I opted to book an online appointment at the local branch, as much to transfer the inconvenience back to the organisation that had created it.

A bank employee then sat with me for roughly twenty minutes to satisfy compliance obligations. The interaction began with a simple question: “Why are you here?” Anyone who has spent time around agentic AI will recognise immediately this. It was textbook intent capture. The autonomous systems we are talking about today will be able to ask that question, instantly classify the request, and route the workflow without ever needing a chair. Instead, the process unfolded manually.

I explained that I was unsure of my CRN. I then provided basic identifying information. Then produced my passport. It was taken away and photocopied. I have submitted digital passport scans for foreign visa applications for years. Entire governments verify identity at national scale using digital processes. Here, a human inspected the document and made a copy.

I then signed a paper form. I updated my details, speaking while he typed. He retrieved my CRN and handed it to me. He then triggered a one-time password so I could re-enter and verify the same information inside my banking app. I typed it all again. Job done.

We can all agree that nothing in that interaction required complex human judgement. It was structured, sequential and deterministic. It was, in effect, an agentic workflow executed by people because the underlying identity data and orchestration layer were not sufficiently trusted to run autonomously1.

Now multiply that experience nationwide. If one million customers require similar service, and each interaction consumes twenty minutes, that equates to twenty million minutes. For a single institution. More than 330,000 hours. Roughly 160 full-time staff years. That’s the hidden cost of identity fragility.

Regulation can force the clean-up. But it is data inconsistency and orchestration immaturity that determine how painful that clean-up becomes. The regulatory stick is not creating the problem. It is exposing decades of accumulated identity drift. The institutions that complete this remediation properly will not simply satisfy compliance. They will harden their operating core and in doing so, unlock nonlinear gains. Because once verified identity is embedded at platform level, every workflow that depends on it becomes safer to automate, every delegated decision requires less oversight and every downstream interaction inherits trust.

The remediation cost, though significant, is incurred once. But the leverage compounds everywhere. Verified identity is not incremental improvement. It is an enabling condition for automation, delegation, and machine-scale trust.

If that is what identity remediation unlocks in banking, it is worth asking what the equivalent looks like in local government. Across Australia, the council Name and Address Register (NAR) underpins almost everything. Rates and property systems. Planning applications. Waste services. Animal registrations. Libraries. Customer request management. Infringements. Community services. It is the quiet spine of council operations. Yet in many councils, the NAR, or more accurately, the multiple NARs, sit in quiet disrepair.

Duplicate ratepayers. Historic land titles still attached to prior owners. Household members fragmented across systems. CRM records that do not reconcile with property data. Decades of manual overrides. Inconsistent address formats across applications. None of these are edge cases. They are structural artefacts of systems built for billing and statutory record-keeping, not for orchestration.

And yet there is increasing pressure for councils to now deliver seamless portals and real-time service coordination, better automate their workflows, and, if you don’t mind, now adopt AI-enabled assistance. Many voices in the sector speak confidently about digital experience yet far less about whether the identity spine beneath that experience is coherent, trusted, and platform-ready.

Whereas banks are being forced to confront identity integrity through regulatory mandate. Councils are not, which I think raises a harder question. In the absence of sector-wide enforcement, does identity remediation simply fall down the priority list? In my own experience, compliance budgets are approved quickly. Foundational data hygiene projects rarely are.

This matters more than ever now because Agentic AI and platform orchestration are not limited by intelligence. They are limited by data confidence. So without a validated citizen and property spine, automation just scales inconsistency. An AI assistant trained on duplicated identities does not fix the duplication. It accelerates the confusion and drives the customer into the front counter. A portal layered over fragmented records does not create seamless experience. It just masks fragmentation behind a digital façade.

And it’s not that it is not possible to achieve better right-sized outcomes. And the ambition for better citizen experience is definitely real. And the technology exists. And the sector talks confidently about transformation. But the NAR remains an unresolved architectural dependency. And this is where the uncomfortable truth sits.

Council CIOs and IT Managers cannot solve this alone. They can propose remediation programs. They can modernise systems. They can integrate platforms. But identity remediation at sector scale is not an IT project. It is a governance mandate.

Banks are not just cleaning millions of records because their CIOs and CTOs were ambitious, though of course many are. They are cleaning them because regulators made identity integrity a board-level risk. So until the civic identity spine is treated as a governance and risk issue, and not merely a system hygiene task, progress will remain incremental. So how do we move this forward?

If we want faster digital progress in local government, we need to reframe the Name and Address Register as critical infrastructure. Not IT. Not a back-office dataset. Not a legacy inconvenience. Critical civic infrastructure.

Councils understand assets better than almost any tier of government. They manage roads, bridges, stormwater systems, community buildings, parks and fleets through disciplined asset management frameworks. They track condition, depreciation, lifecycle risk and renewal schedules. They allocate capital to prevent failure. They plan decades ahead. Local government is, at its core, an asset portfolio manager. Now imagine if that same legislative discipline were applied to civic identity data.

It’s not such a bold step. Councils exist through state legislation. In most jurisdictions, asset management planning is mandated. They are required to demonstrate lifecycle planning, risk mitigation, renewal forecasting and financial sustainability across their physical asset base. The governance machinery already exists.

Imagine a state-level directive requiring councils to validate, reconcile and standardise their Name and Address Register over a defined period. Standardised address formats. Duplicate detection and remediation. Cross-system identity reconciliation. Clear ownership. Auditability. Executive reporting on data integrity metrics. Not as an IT uplift. As mandated asset stewardship.

It would not be glamorous. It would not generate headlines the way KYC has in banking (though it probably should). But it would create the conditions under which digital ambition could safely operate.

State governments already exercise oversight of councils in matters of financial sustainability and infrastructure risk. Extending that discipline to digital identity would not be radical. It would simply recognise that the civic identity spine now underpins service delivery as materially as roads and stormwater. The sector does not need another transformation slogan. It just needs the Name and Address Register treated with the same seriousness as a bridge.

Achieve that, and its even possible to see a little further down the orad. Across Australia, there is frequent discussion about councils sharing the cost of technology platforms. Shared ERP environments. Shared CRM. Regional system consolidation. The logic is understandable. Pooling investment appears efficient. Buying the same system sounds rational. And in principle, it is not a bad idea. But it is incomplete because systems do not create capability. Data does.

Two councils running the same platform with inconsistent, duplicated and poorly reconciled Name and Address Registers will not achieve interoperability. They will share infrastructure, but not integrity. The argument to “buy the same system” is easier to grasp. It is concrete. It can be tendered. It can be budgeted. It produces visible alignment.

But shared data discipline is harder. It requires governance alignment, common standards, reconciliation effort and executive mandate. It does not generate ribbon-cutting moments. But in practice, the capability of any shared system is determined by the quality of the data flowing through it.

A shared platform built on fragmented identity registers simply aggregates fragmentation at scale. If we are serious about shared technology services, then the logical precursor is shared identity integrity. Otherwise we are optimising the container while ignoring its contents.

What will not go away is the fact that, as citizens, we will constantly demand better digital engagement with our local authorities. Every new generation will expect faster responses. Better portals (and agents!). Coordinated services. Those experiences are built on identity integrity.

“Our” banks know that regulatory pressure accelerates the strengthening of that foundation. Our local governments probably do too and should not wait for a similar stick.

If councils are serious about AI, orchestration and modern digital experience, the transition must be led from the top, framed not as digital uplift, but as governance and risk management. Banking was compelled by KYC. Not because it was technologically elegant, but because it was institutionally undeniable. Boards understood it. Audit committees tracked it. Executives funded it.

Local government needs its own forcing function. One that reframes the Name and Address Register as critical civic infrastructure. And one that translates identity integrity into the language of risk and stewardship that non-technical leaders already understand, much as KYC did in banking.

Without that shift, digital ambition will roam the streets of an LGA near you. Visible, energetic, and politically attractive. All while the identity infrastructure it relies on remains untreated.

Everyone is racing toward an AI-enabled future state, but most organisations are working on the wrong problem first. The real challenge of 2026 is not model capability, agent maturity, or responsible AI frameworks. It is the far harder work of changing the ERP-centred operating model that still sits underneath almost every enterprise. AI will arrive regardless. The question is whether organisations will be structurally ready for it. And who is accountable for making them so.

For decades, ERP has been the gravitational centre of enterprise technology. Not just as a system of record, but as the implicit organiser of work, authority, and accountability. Processes were designed around its constraints. Humans filled the gaps. Control lived in people as much as in systems. That model worked until intelligence began to move into the machine.

AI does not fit neatly into an ERP-centred world. It does not tolerate implicit intent, fragmented state, or human-held orchestration. When dropped into these environments, AI does not transform them, it compensates for them. So humans remain in the loop not because judgement is required, but because the operating model cannot yet stand on its own.

That is why the real AI task in front of us is architectural, not algorithmic. To reach a future where AI can act, not just advise, organisations must first decentre ERP as the organising principle of work and reframe it as a still critical but bounded system of record within a broader platform architecture. This is not about replacing ERP. It is about relieving it of a role it was never designed to play.

The inevitable AI future state cannot be achieved by layering more intelligence on top of yesterday’s operating model. It can be achieved by doing the harder, less glamorous work of redesigning how intent, flow, and authority are expressed across the enterprise. That is the real AI challenge of 2026. So let’s get busy changing the world.

One of the quiet failures of 21st-century enterprise technology so far is that we keep using 20th-century language to describe it. “Human-in-the-loop” is the latest perfect example. It is usually presented as wisdom. A familiar sign that an organisation understands the limits of automation and prudently respects the role of its people. In reality, I think it really signals something far less noble.

When AI strategy or product selection depends on humans remaining in the loop to make the system work, the problem isn’t ethics or caution, it’s architecture. What you’ve really done is extend an ERP operating model, wrapped it in AI, and moved it onto modern infrastructure.

AI has exposed the limits of application sprawl, outsourced responsibility, and ERP-centred thinking, all relics of a 20th-century operating model that no longer holds. The irony is that the one construct we should have preserved to manage this complexity, the CIO as a true technology leader, has been steadily dismantled at the very moment it is most needed.

For most of the modern enterprise era, systems were built around transactions, not flow. Work happened because a human initiated it, pushed it forward, interpreted exceptions, and absorbed ambiguity. Technology recorded outcomes. But people carried the intent of the transaction. ERP systems formalised this worldview.

Byut they were never designed to run organisations autonomously. They were designed to support human-led execution. Roles, approvals, screens, handoffs, and controls all assumed that a person sat at the centre of the loop.

When AI is layered onto this environment, it does exactly what you would expect. It becomes expontentially smarter at the edges. It summarises, explains, recommends, and reassures better than we can. It improves interaction without changing the underlying structure. So humans stay in the loop. They approve what the system cannot approve. They escalate what the system cannot interpret. They stitch together what the architecture cannot orchestrate.

This becomes a serious problem at scale. Boards, whether public or private, exist to govern systems that must operate reliably, repeatedly, and under pressure. Not once. Not in a pilot. Not in a demo. At national or enterprise scale, with failure modes that are political, legal, financial, and human. We’ve seen this before as part of every significant 21st century change.

Rolling out a national vaccination program is not a communications exercise. It is a system-of-systems problem involving supply chains, identity, eligibility rules, appointment scheduling, exception handling, adverse event reporting, and public trust, all operating under intense scrutiny and time pressure. Boards were not asking whether the intent was ethical. They were asking whether the system could withstand volume, variation, and failure without collapsing.

The same is true of the shift to digital driver licences. Once a digital licence becomes a primary credential, it must work everywhere, all the time. Offline, across jurisdictions, under enforcement, and in edge cases the designers never anticipated. A board does not care that the user experience is elegant if the system fails during a roadside stop, a disaster response, or a court proceeding. Reliability and accountability are non-negotiable.

Or take the current nationwide KYC and identity reforms in Australia. These are not digital initiatives. They are foundational trust infrastructures. When onboarding fails, payments stall, benefits are delayed, or fraud scales faster than control, the consequences are immediate. Bank boards are not reassured by good intentions. They want to know where authority sits, how decisions are enforced, and who is accountable when the system makes, or enables, a mistake.

In all of these cases, humans are involved, but critically, not because the system would otherwise fall apart. They are involved because judgement, discretion, and oversight are genuinely required at the edges. The core system is designed to stand on its own. Human involvement enhances it; it does not prop it up. That is the standard boards apply. Which is why this matters so much in AI discussions.

When AI-enabled systems are presented to boards with the reassurance that humans are in the loop, but without clarity on whether those humans are adding judgement or simply preventing failure, boards are being asked to underwrite risk without being shown the architecture. That is not how large-scale systems are governed. Boards understand the difference between human oversight and human dependency. One is a strength. The other is a liability. But the distinction is not always made explicit in how AI is presented to them. And at scale, that distinction is everything.

At scale, boards care less about intent than about exposure. When something goes wrong, they actively look for where control actually sat and who carried the consequences. If a system cannot function without humans, accountability does not sit with the technology, it sits with the people. And boards rarely tolerate hidden accountability.

Hidden accountability is risk. Diffuse accountability is risk. Implicit accountability is risk. This is why failing technology initiatives so often become personal at board level. Failure exposes whether control ever truly existed. Boards do not fund virtue. They fund capability, control, and outcomes.

On the factory or the trading floor, responsibility has very little to do with whether a system needs human intervention to function. Architecture does. And so this brings us to where platforms fundamentally break with ERP thinking.

A real platform is not defined by UI, licensing model, or cloud credentials. It is defined by where control lives. In a platform environment like ServiceNow, control is designed to live in the architecture, not in people’s heads or inboxes. Workflow is explicit, not implicit. State is durable. Authority is delegated, not implied. Policy is externalised from process. Integration is transactional and observable. The truth only lies at a modular level.

Every serious platform converges on the same capabilities, and notably, these are platform-as-a-service constructs, not ERP ones. Explicit workflow orchestration (e.g. Servicenow’s Flow Designer). Transactional integration (e.g. IntegrationHub). A trusted model of operational reality (e.g. CMDB). Enforceable policy and security controls. Not because vendors agree, but because autonomy demands that policy and decision logic exist so rules stop being folklore and start being code. Without these foundations, AI can reason, but it cannot be trusted to act.

These are not additional products or services to be included in an ERP roadmap. They are the structural foundations on which autonomy depends. And when they are absent or immature, humans must step in to compensate. Humans become the workflow engine. Humans become the integration layer. Humans become the permission model. Humans become the exception handler.

The real outcome is that humans in the loop has never been about man-machine collaboration. It’s a cover for unpaid architectural debt. And this is why so many early conversations about agentic AI have felt underwhelming.

The agent can look and sound intelligent. It can analyse, categorise, plan, and make suggestions. But when it comes time to act, it often can’t. The ERP-centric environment it lives in doesn’t give it real authority, clear boundaries, or end-to-end accountability. So the work stops at the screen. A human has to step in and finish the job. The demo looks impressive, but the client operating model underneath hasn’t actually changed.

For much of last year this was explained away as ethical restraint, and while it is true that reality is never black and white, in many cases, that language is masking a more uncomfortable truth. The architecture (of some solutions) and many organisations simply isn’t ready yet.

When presented to customers by a technology provider early in their AI maturity curve, human-in-the-loop is not a sign of wisdom. It’s a sign of platform limitation. Only later, when the system can operate reliably on its own, as it adopts more PaaS-centric models, does keeping humans in the loop become a genuine design choice, and a source of strength.

In immature environments, humans are in the loop mostly because the system cannot be trusted. In mature environments, humans are in the loop because their judgement is genuinely valuable. Those two states look identical on a governance slide whereas they could not be more different in reality. Most of the high-profile failures of 2025 sit squarely in that gap when organisations convinced themselves they were in the second state while still operating firmly in the first.

One means humans are propping up brittle systems or solutions. The other means humans are governing autonomous ones. Which brings me back to the beginning. The entire purpose of a platform is to enable the once in a generation transition wea re living through in real time.

This is not about removing humans. It never was. It is about relocating humans to where 21st-century organisations actually need them. Not as workflow glue. Not as integration buffers. Not as permission proxies. But as stewards of intent, ethics, accountability, and outcomes.

When platforms mature, humans move out of execution and into governance. They become orchestrators. They stop making the system work and start deciding how it should behave. That is also when AI stops being dangerous. Not because it is constrained, but because it can be contained by architectural design.

The opportunity is to move beyond an ERP-centric operating model, not to abandon ERP altogether. ERP isn’t going away. But it is being moved out of the centre, because organisations are now dealing with a genuine three-body problem between ERP, PaaS, and AI. Each exerts force on the others, and none can dominate without creating instability. ERP’s role is to remain a system of record, not the system of control. The centre of gravity is shifting toward platforms that can express intent, orchestrate work, and delegate authority across the estate, with AI operating inside those boundaries.

Regardless of which industry you operate in, crossing that threshold does not start with more AI. It starts with an honest reassessment of the operating model itself and a willingness to accept that the future of ERP is as a critical component of the value chain, but not its organising principle. That is not a loss of control. It is how control is regained for the next era and it should rewrite most procurements from this year forward.

Enterprise technology buyers are being told a comforting story. Artificial intelligence, we are assured, can be bought, licensed, and governed much like the systems that came before it. Enterprise licence agreements, flexible consumption pools, and familiar commercial constructs promise to make AI feel legible and safe. Salesforce’s Agentforce Enterprise Licence Agreement (AELA) is a good example of this intent. It is not a gimmick. It is a solid and sincere attempt to meet customers where they are.

The problem is not that these models are wrong. They are right for now. But that is also precisely the issue. They are transitional. They wrap a fundamentally different economic creature in the language of a world that no longer exists. Traditional total cost of ownership (TCO) was built for software that behaves like an asset. Agentic systems behave like labour. And that distinction matters more than any pricing model.

For decades, TCO has been an exercise in containment. Organisations estimated licence costs, infrastructure, integration, and support, then spread that cost across users or transactions. Performance improvements were welcomed because they reduced run costs. Efficiency meant savings. Usage broadly tracked value, and behaviour was bounded by human attention, working hours, and organisational friction. Agentic systems break every one of those assumptions. How and why?

An agent does not wait to be used. It acts. It retries. It escalates. It triggers other systems. It improves over time and continuously operates. Cost is no longer driven by access but by behaviour. In traditional systems, better performance was efficiency driven to reduce cost. Whereas in agentic systems, the more capable an agent becomes, the more it does. So better performance often increases use. This is where familiar TCO models quietly fail.

An organisation may deploy an agent to reduce case handling time, improve resolution rates, or increase customer satisfaction. All of those outcomes can be achieved. Yet the same improvements can also drive higher system interaction, greater orchestration complexity, more monitoring, and increased downstream activity. The agent becomes productive, but not necessarily cheap.

That creates the new TCO question. Are we paying for fewer actions or for better outcomes? The difficulty is that even this framing assumes a level of comparability that rarely exists. I’m sure the model enthusiasts reading along have already quietly objected to the efficiency argument.

But I agree. Two organisations can deploy the same agentic process, on the same platform, under the same commercial model, and experience materially different economics. One may operate with clean data, minimal workarounds, and low regulatory friction. Another may require multiple exception paths, manual validations, and heavy oversight. The agent performs the same “work,” but with very different efficiency, risk, and cost profiles. That doesn’t change the argument that traditional TCO has no way to account for this because it was designed to price systems, not organisational complexity.

This is why agentic cost benchmarking remains elusive. What looks like an AI efficiency problem is as much an operating-model problem in disguise. That is a management consulting problem and the result is the ensuing economic fog.

Customers want AI. Boards expect it. Executives feel pressure to adopt it. Yet few organisations truly understand how to model its long-term cost, let alone govern it. Everyone is struggling to determine whether to buy capacity, consumption, outcomes, or something in between. The language of tokens, actions, and autonomy feels abstract and risky.

To this point vendors have only been able to respond rationally. They have to offer what sounds safe and familiar. Enterprise licence agreements. Flex pools. Commitments that resemble what procurement teams already know how to approve. These constructs are not cynical. They are necessary bridges. They allow organisations to step into agentic territory without immediately confronting the fact that the ground rules have changed. But the ground rules have changed.

In many ways, this challenge should feel familiar. It is the same problem organisations have wrestled with for decades in business process outsourcing and managed services. Cost outcomes were never determined solely by the contract rate card. They were shaped by process maturity, regulatory burden, exception volume, and the level of oversight the organisation itself required. Two customers could outsource the same process and experience radically different economics. Agentic TCO brings that same reality into software.

At this point, it is important to be precise. Not every layer of an AI-enabled platform behaves the same economically. Foundational PaaS capabilities like workflow engines, data models, and integration layers can still be governed and amortised like traditional software assets. Their cost curves are familiar, and their value can be planned. The discontinuity appears at the point where agency is introduced. Once software is empowered to act, decide, and initiate work, cost becomes behaviour-driven rather than asset-based. This is why vendors are carving out separate commercial constructs for agentic capability. It is not an admission that platform pricing is broken, but an acknowledgement that agency cannot be priced in the same way as software.

Agentic TCO is not about ownership. It is about economic governance. It shifts the focus from controlling access to controlling autonomy, from budgeting per user to budgeting per action, and from minimising cost to constraining behaviour. That shifts the dominant cost drivers from licences and infrastructure to decision loops, orchestration logic, exception handling, guardrails, and oversight. These become operating costs, not implementation artefacts, and they tend to grow with success rather than diminish over time.

Let me beat my old drum again. This is why platform architecture matters. Agentic systems do not sit neatly inside application silos. They require a platform layer that can observe, govern, throttle, and evolve behaviour across systems. This is not an ERP problem. It is a platform-as-a-service problem. PaaS becomes the economic trunk through which agentic activity flows, whether organisations acknowledge it or not.

So what we are witnessing is a market in transition. Customers are being sold AI using commercial models designed to reduce anxiety, while vendors quietly pivot toward a future where agency, not software, is the unit of value. Enterprise agreements like Salesforce’s AELA are part of that necessary bridge. They help organisations begin the journey by uncapping some of that agency. But even they will admit that they do not yet resolve the underlying economic ambiguity at free agency scale. No one has.

The bottom line is that agentic AI cannot be priced and governed like software. It must be governed the way work is. And until organisations accept that shift, total cost of ownership will remain an estimate rather than an insight. Is it still worth pursuing? Absolutely! The shift is simply this. Where traditional TCO asked what a system costs to own, agentic TCO asks what it costs to let software act on your behalf.

I think we all get that part of the discomfort to this point comes from a reluctance to talk about AI taking jobs. Framed that way, the conversation becomes politically charged and emotionally loaded. In practice, agentic AI is not taking jobs. It is taking work. The tasks and decisions that were previously performed by people. That distinction matters, because work can be delegated, reallocated, and governed in ways that jobs cannot.

Agentic AI is not difficult to price because vendors lack models, but because enterprises have not yet accepted that they are buying delegated work rather than tools. So try reframing the way you are approaching those business cases. Everything else is just a negotiation over how long we pretend those two questions are the same.

Share

The Top 10 Questions to Start Costing Agentic AI

Costing agentic AI is not difficult because the technology is new, or because vendors lack pricing models. It is difficult because organisations are trying to apply asset-based thinking to something that behaves like delegated work.

Traditional TCO models assume comparability, predictability, and efficiency gains that reduce cost over time. Agentic systems violate those assumptions. Their costs are shaped as much by organisational complexity, governance choices, and operating discipline as by software consumption itself.

The questions below are not designed to produce a perfect number. They are designed to surface where agentic AI will behave unlike traditional software, where costs will scale with success, and where economic exposure is likely to appear. Without answering them, any TCO model will be incomplete, regardless of how familiar the commercial wrapper looks.

They focus on the agentic (probablistic and variable) layer of the platform, not the underlying PaaS foundations, which can still be governed and amortised using traditional asset-based TCO models. They are based on the premises that you can’t cost what you can’t see, you can’t budget what you can’t limit, and you can’t govern delegated work without ownership.

At scale, organisations are likely to require a control layer that can observe, constrain, and govern agentic behaviour across systems. Such control mechanisms should not be viewed as a product trend, but as a structural response to the economic realities of delegated work operating at machine scale.

Read more

Globalisation is a relatively recent phenomenon. And in the early 2000s, something subtle but consequential began to unfold inside Western organisations. Standing in line for coffee in the Sydney CBD, shoulder to shoulder with the bankers of George Street and Martin Place, it was easy to assume that everything was working as designed. The offices were full. The deals were flowing. The machinery of capitalism looked solid.

But what began as a quiet experiment in cost reduction soon exposed a far deeper truth. When organisations attempted to move work beyond their walls, they discovered they didn’t actually understand how that work got done. Processes lived in people’s heads. Decisions relied on proximity. Exceptions were handled by habit rather than design. Basically, the first wave of globalisation forced organisations to confront the fact that they didn’t truly know themselves.

Business Process Outsourcing and Offshoring was supposed to be about labour arbitrage. Lower wages. Time zone leverage and scale. Large banks were among the earliest adopters, not out of experimentation but necessity. When margins are defended at the second decimal place and operational efficiency compounds at scale, even small reductions in unit cost become strategically meaningful.

What it became instead was a forced confrontation with organisational reality. Processes that lived in people’s heads had to be written down. Exceptions had to be named. Decisions had to be justified. Handoffs had to be explained to someone who wasn’t sitting three desks away. And for a while, progress was slow. Painfully slow.

Executives underestimated the sheer volume of documentation required. BPM diagrams multiplied. SOPs sprawled. Edge cases turned out to be the rule rather than the exception. And there was a deeper discomfort too. Not just with process discipline, but with trust. Trusting distant teams. Trusting unfamiliar cultures. Trusting companies most leaders had never even heard of.

Even as late as 2005-06, there was still a surprising lack of clarity about what was unfolding. At IBM’s Australia–New Zealand Sales Kick-Off on the Gold Coast, I was asked to stand in front of more than 300 sales and account executives and explain who the “Indian outsourcers” were and why they mattered. But then, almost without warning, the floodgates opened.

Once processes were explicit, portable, and measurable, they stopped being local. Work that had once been anchored to physical offices began to flow. Entire delivery models shifted. Cities like Bangalore, Pune, and Hyderabad didn’t rise because of cheap labour alone. They rose because execution itself had become transportable.

The result was profound. A generation of global technology firms emerged. Economies rebalanced. New talent ecosystems formed. The centre of gravity shifted and it never fully shifted back.

I saw that transition up close. In 2004, I travelled to Manila to close the local META Group office near Makati City. At the time, Bonifacio Global City literally did not exist. Other than in developer brochures. The land between Makati and the old military reservation was sparse, quiet, and largely undeveloped. It was a place passed by, not a place you went to.

Less than a decade later, that same ground was unrecognisable. Where there had been open land, there were now dense clusters of glass and steel high-rise offices, residential towers, and global delivery centres operating around the clock. Bonifacio had become a city in its own right and one of Asia’s most important hubs for global outsourced execution. That is the scale of change BPO set in motion.

I’ve been thinking a lot about that period lately and am increasingly convinced that while the delivery platform has changed, we are otherwise watching the same tectonic shift occur all over again.

Last time, the logic of optimisation drove execution to the only places where cheap labour at scale still existed, and entire cities rose to serve that demand. This time, the same logic applies, but the city won’t be physical.

Agentic AI is currently in its “this looks promising but feels messy” phase. A lot of early pilots have stalled. Proofs of concept have struggled to scale. Leaders have complained that the tools are impressive but unreliable. Teams have quietly discovered that their processes aren’t nearly as clean as they thought they were. Decisions that felt obvious have turned out to be contextual. Exceptions have exploded.

I think the disconnect is less about technology and more about readiness. Many vendors are aggressively marketing Level 5–6 agentic capabilities to organisations that are still struggling to define their Level 1–2 foundations. The result is that sales and delivery organisations end up aligned to the pace of marketing rather than the maturity of the customer. And while the two can coexist, operations cannot move at the speed of marketing. That was never the intent.

But this hasn’t been a failure of AI. It’s exactly the same organisational reckoning we saw during the first BPO and offshoring wave. In the same way that you can’t outsource chaos, you can’t delegate incoherence to an agent.

Before anything meaningful can happen, work has to be made legible. Inputs defined. Outcomes bounded. Authority clarified. Escalation paths designed. Controls articulated. Metrics agreed. The real work, once again, isn’t automation. It’s self-knowledge. That’s something we’re all impatient for.

That’s why progress feels slower than the headlines suggest. The technology has been ahead of the organisations adopting it. Exactly as it was twenty years ago. But history suggests this phase doesn’t last forever.

That was the hidden force behind offshoring. Documentation didn’t just enable delivery. It shifted the centre of execution. Once work could move, it did. Talent followed. Capital followed. Capability clustered.

Agentic AI creates the same condition, but with a far more radical implication. Execution itself is becoming abstract. That is not hyperbolic. Let’s ponder that for a minute. Agents literally don’t sit in buildings. They don’t sleep. They don’t resign. They don’t require visas or commute times. They consume tokens, operate within guardrails, and escalate when designed to do so. They can also be insourced, outsourced, federated, shared, or rented. Geography becomes optional. Time becomes elastic.

But this is also where the analogy with BPO stops being clever and starts being uncomfortable. In the 2000s, organisations had to decide whether to build captive centres offshore or outsource to third parties. Some did both.

Today, the choice reappears in a new form. Do you run agents internally, tightly coupled to your data, architecture, and governance? Or do you consume agentic capacity as a service, trusting external platforms to execute parts of your organisation on your behalf? If the decision feels tactical, that is a mistake. It isn’t. Because just as before, the early choices will harden into operating models that last decades.

Contemporary BPO cities and offshore delivery hubs came into existence because of density. Skills, services, capital, and coordination concentrated until something self-reinforcing emerged. Once enough capability clustered, those cities stopped being back offices and began to function as innovation engines.

Over time, some went further still. As talent deepened and decision-making followed execution, these centres began to assume strategic importance, shaping roadmaps, influencing investment, and increasingly determining where growth would come from. What started as delivery hubs evolved into centres of gravity in their own right. We’re seeing this occur today in the way large global technology vendors are balancing India as a strategic execution and innovation centre, with Singapore increasingly acting as a regional coordination and administrative hub.

Agentic ecosystems are already showing the same tendencies. Common agent frameworks. Shared orchestration layers. Specialised agent roles. Reusable process patterns. Governance primitives. Observability tooling. Token optimisation. Cost-to-serve models. These are the streets, utilities, and institutions of a new kind of city.

It won’t have a skyline. It won’t appear on a map. But it will concentrate execution, capability, and influence in ways that are no less real for being invisible. We’re already seeing early versions of this in the wild. Platforms like ServiceNow describe it as a control tower. The place where work, agents, workflows, and decisions are orchestrated rather than directly executed. Accenture talks about agent-led delivery factories embedded inside client operating models. IBM, drawing on decades of systems integration and automation, frames it as cognitive operations.

The language differs, but the struggle is the same. That is, to become the de facto regulatory authority for digital workforces setting the rules, enforcing the controls, and deciding how autonomous execution is allowed to become.

Digital agentic cities won’t live entirely in the cloud or inside legacy systems. They will span them coordinating execution across environments that were never designed to be seen as a single place. For that to work, they must be interoperable by design and capable of managing multiple digital labour forces within a single, regulated operating environment. Not one workforce, not one platform, but many (human, automated, and agentic) governed together inside the complexity of a contemporary enterprise.

Organisations that align themselves early will find execution cheaper, faster, and more adaptable. Those that don’t will struggle to compete with firms whose workforce can scale at machine speed. The uncomfortable truth is that these cities are being built regardless, and without a central authority, what emerges isn’t innovation, but sprawl.

Share

Much of the public debate about AI has to this point fixated on human displacement. That’s understandable, but I also think it misses the deeper pattern.

BPO didn’t eliminate work. It reorganised it. Yes, it did hollow out organisations that couldn’t define themselves clearly but it also elevated those that could orchestrate complexity across distance. Agentic AI will do the same.

It will punish organisations that rely on tacit knowledge, heroics, and informal coordination. It will expose unclear decision rights. It will surface governance gaps that have been tolerated for years. It will make ambiguity very expensive. But it will also reward those that understand their own architecture, both technical and organisational.

This is why AI strategies that start with tools and finish with dashboards feel fundamentally incomplete. The challenge isn’t intelligence or automation. It’s whether an organisation can describe itself clearly enough to be executed by something other than habit. That’s why this is a leadership moment. CEOs don’t need to understand AI, but they absolutely need to understand the organisational consequences of deploying it.

What is undeniable is that the offshoring boom reshaped global technology dominance because it changed where execution lived. Agentic AI will reshape it again by changing what execution is.

The slow burn we’re seeing now, the frustration, the false starts, the documentation fatigue, is not a sign that this moment is overhyped. It’s a sign that we’re still early. Confidence should come though from the knowledge we have lived through it before. And we came out stronger. For those early in their careers, that means experiencing a period of uncertainty where the rules are shifting faster than the narratives explaining them. Because the very shape of work is changing underneath us all.

But once the legibility problem is solved at scale, and once agents become trusted participants in organisational workflows rather than novelties, the shift will feel sudden. It always does. And by the time it’s obvious to everyone, the agentic city will already exist.

And perhaps that is the quiet unease sitting underneath all of this. Not that we are entering a simulation, but that we are finally seeing how simulated our organisations already were. Roles, approvals, escalation paths, and incentives are all abstractions we agreed to treat as real because humans were inside them.

Agentic AI doesn’t invent a new world. It just removes the camouflage that humna-in-the-loop has always provided. And when execution no longer requires presence, belief, or habit, what remains is structure. Some organisations will recognise themselves in that reflection. Others will discover that what they thought was a living system was mostly theatre.

The city of digital agents will keep growing either way. The only real questions left will be who will govern it, whether you helped build it, and whether your work quietly moved there without you.

Above all else, there is a quiet irony here. The captive centres and offshore providers that industrialised execution during the BPO era are likely to be among the first organisations forced to confront an agentic future. I don’t think it is a case that what made them indispensable last time makes them replaceable first this time.

Rather, after spending two decades breaking human work into repeatable systems and delivering it at scale, it is that they now find themselves confronting the same optimisation logic they once helped their clients apply, only this time where that same work can be executed by something else entirely.

For them, the change may be as great, where the implications of agentic execution are not theoretical. They are immediate, unavoidable, and impossible to ignore. Paid subscribers can read on for a closer look at the ten Australian organisations and five technology vendors I’ll be watching in 2026 as they shape the early contours of the agentic decade.

The transition to agentic execution won’t be evenly distributed. It never is.

In Australia, the first organisations to feel and shape this shift will be those that already operate at scale across banking, financial services, insurance, and telecommunications. These sectors didn’t just adopt offshoring early. They normalised execution as an industrial discipline. That makes them the earliest proving ground for agentic control.

The 10 Australian companies (and their competitors) I’ll be watching closely in 2026 combine large captive or offshore workforces, high process maturity, and relentless optimisation pressure. They are where agentic execution stops being theoretical.

1. ANZ: One of the largest Australian captive footprints across Bengaluru and Manila, embedded deep into core operations and technology.

2. Commonwealth Bank of Australia: Thousands of offshore staff across engineering, operations, and change, with unusually high execution discipline.

3. National Australia Bank: Long-standing offshore delivery and innovation model now colliding with platform and automation pressure.

4. Westpac: Mature offshore execution fabric supporting core banking and transformation.

5. Macquarie Group: Globally distributed services and operations where execution fluency already drives strategy.

6. Suncorp Group: Insurance and banking execution at scale, increasingly standardised and automatable.

7. Telstra: Large India-based capability and innovation centres underpinning network, digital, and service operations.

8. Optus: Heavy reliance on offshore execution across customer and network domains.

9. Qantas: Complex, safety-critical execution environments with growing automation and offshore dependence.

10. Woolworths Group: Less obvious, but enormous scale across supply chain, analytics, and digital operations where execution is already highly systematised.

These organisations won’t debate whether agentic execution matters. They will be forced to decide how it is governed.

Then there are the 5 vendors I’ll be watching that are competing to define agentic control and influence how that authority is established. As I covered in the main article, the real contest is not about who builds the smartest agents. It’s about who becomes the independent regulator for mixed human and digital labour.

1. ServiceNow: Explicitly positioning itself as the control tower for work, agents, workflows, and decisions across the enterprise.

2. Accenture: Embedding agent-led delivery factories directly into client operating models, shaping execution from the inside.

3. IBM: Leveraging mainframe, hybrid cloud, and decades of systems governance to frame agentic execution as cognitive operations.

4. Microsoft: Quietly defining who and what is allowed to act through identity, security, data, and platform control.

5. Salesforce: Extending from vertically intergrated customer systems into broader orchestration of enterprise work and agents.

When the most disciplined execution environments in these ecosystems begin to move, the rest of the market will not be far behind.

And a final note. This analysis does not cover enterprises pursuing a predominantly internal build strategy, where agentic capability is treated as core infrastructure rather than an external service. That’s an approach that while it does include these vendors, is also increasingly associated with platforms such as Glean and Kore.ai.