The 2026 State of Agility in Procurement & Supply points to a widening gap between the speed of technological capability and the speed at which organisations can actually turn that capability into value. For technology leaders, the critical question is shifting from what AI can do to whether the architecture, data, governance and operating model around it are capable of moving at the same pace.
Artificial intelligence is advancing faster than most enterprise operating models.
Analytics are improving. Automation is becoming more capable. Generative and agentic systems are compressing tasks that previously required significant manual effort. Decision-makers can increasingly access insight in seconds rather than days.
Yet the 2026 State of Agility in Procurement & Supply suggests that technology capability is no longer the main constraint.
Seventy per cent of respondents believe their current operating model constrains the benefits that could be achieved from AI. Among respondents who have adopted AI, 73% report that adapting the operating model required to scale those benefits remains difficult or only partially implemented.
For CIOs, CTOs and digital leaders, that changes the transformation agenda.
The challenge is not simply whether the organisation has the right models, platforms, cloud environment or automation stack. It is whether information, authority, governance and work can move across the enterprise in a way that allows those technologies to create end-to-end value.
Technology performance is becoming system performance
Enterprise technology has traditionally been measured through a familiar set of lenses: reliability, security, cost, integration, user experience, data quality and delivery performance.
Those remain essential. But the State of Agility research points towards a broader question: how effectively does the organisation itself operate once technology has made individual activities faster?
The report shows that 96% of respondents consider business agility strategically important in dealing with greater market uncertainty. At the same time, 94% say functional silos limit or delay value creation to some degree. More than half, 56.54%, describe that effect as significant and another 9.42% as critical.
That combination should matter to technology leaders because digital transformation increasingly runs into problems that software cannot resolve on its own.
A workflow can be automated and still wait for approval. A data platform can provide a single source of truth while functions continue to optimise against different objectives. An AI assistant can produce a recommendation instantly, but the organisation can still lack clarity over who has authority to act on it. Integration can move information across systems while governance continues to move decisions through sequential queues.
The largest bottlenecks reported in the study reinforce that point. Requirement and scope definition emerges as the biggest source of delay, followed by governance and approvals and then finance approvals and budgeting. The leading root causes behind organisational slowdown are conflicting functional objectives, limited transparency between silos and changing requirements or constraints.
These are not primarily infrastructure problems.
They are operating-model problems expressed through technology.
That distinction is increasingly important because technology teams are often asked to solve friction that originates elsewhere. A new platform is expected to compensate for unclear ownership. Automation is expected to compensate for excessive approvals. Better dashboards are expected to compensate for weak decision rights. AI is expected to compensate for fragmented processes.
In each case, technology can improve the local experience without fixing the structure that created the friction.
The report captures the issue simply: every handoff creates the potential for delay, distortion and loss. AI can improve individual steps. Only redesign improves the system.
For technology leaders, the next phase of transformation is therefore less about digitising the existing organisation and more about deciding which parts of the existing organisation should survive digitisation at all.
The AI test: capability is not the same as operating leverage
The research does not suggest AI investment is failing. Respondents already report clear benefits.
Improved analytics and reporting accounts for 27.59% of reported AI benefit. Automation of operational tasks represents 26.11%, faster business insights 23.65% and self-service enablement 19.7%.
These are meaningful gains. They also reveal where many organisations currently are in the adoption curve.
Most benefits enhance existing work. AI analyses information faster. It automates a task. It surfaces insight sooner. It enables a user to complete an activity without manual support.
The larger opportunity begins when technology leaders ask a different question: what should the operating system of the enterprise look like now that these capabilities exist?
The report shows that traditional organisational structures remain dominant, including divisional and business-unit models, functional structures and matrix organisations. Truly adaptive and AI-enabled operating models remain comparatively uncommon.
That creates a familiar technology paradox. Organisations are deploying tools designed for speed and adaptability into structures built around sequential handoffs, fragmented data ownership, fixed responsibilities and approval-driven governance.
The result is local optimisation.
A procurement workflow becomes faster, but finance approval remains unchanged. Contract analysis becomes faster, but legal enters the process too late. A planning engine improves forecasting, but the supply chain still works against disconnected objectives. A customer platform improves visibility, but data ownership remains contested across functions.
The technology works. The system around it does not move.
That is why architecture can no longer be treated purely as a technical discipline. Increasingly, enterprise architecture and operating-model design are converging.
The research suggests respondents can already see the characteristics of a stronger model. They describe high-performing commercial operating models as data-driven and dynamically adjusted, built around empowered cross-functional teams and increasingly supported by AI-enabled platform teams capable of self-service. Their target state is defined most strongly by cross-functional empowerment, AI-enabled and data-driven decision-making, and agile, adaptive and simplified governance.
Technology is central to that model, but technology is not the model.
The model is the interaction between people, process, data, governance, decision rights and technology.
For CIOs and CTOs, that provides a more demanding test of transformation investment. Does the platform simply improve a function, or does it reduce friction across the value stream? Does AI make an existing approval process faster, or does it allow the organisation to redesign the approval model? Does a new data layer improve reporting, or does it create shared visibility that changes how teams make decisions together?
The difference is the difference between digital efficiency and operating leverage.
Data, integration and governance are now inseparable
The scale of the opportunity becomes clearer when the report examines organisational speed.
Respondents were asked how much faster their organisations could operate if fragmentation and barriers to shared decisions and accountability were reduced. 44.08% estimated a 20% improvement and 38.82% estimated a 50% improvement. A further 9.87% selected between 10% and 20%, while 4.61% believed the organisation could become 100% faster. Only 2.63% expected no improvement.
These are respondent estimates rather than measured productivity outcomes, but they expose how much capacity leaders believe is trapped in organisational friction.
For technology teams, some of that friction is visible in the architecture itself.
Multiple systems hold competing versions of the same information. Data is transferred between functions rather than shared by design. Workflows encode historical approvals that no longer reflect actual risk. Permissions mirror organisational charts rather than value streams. Integration projects connect applications while leaving process ownership fragmented.
This is why data strategy, integration strategy and governance design increasingly need to be considered together.
A shared data layer without shared decision rights can simply make disagreement more visible. Automated workflows without policy redesign can accelerate unnecessary control. AI without trusted data can increase the speed of poor decisions. Governance that remains entirely manual can become the throttle on increasingly automated systems.
The report’s focus on platform-based self-service is therefore significant.
Commercial platform teams emerge as one of the practices respondents associate with new operating models, alongside Lean-Agile Procurement, empowered cross-functional teams and short feedback cycles. The underlying idea is not merely another technology platform. It is a different service model: repeated requests are converted into governed, reusable capabilities that people can access without recreating the same transaction every time.
That is a technology and operating-model decision at the same time.
The strongest digital organisations will increasingly distinguish between work that requires expert judgement and work that can be safely converted into policy, platform and self-service.
That requires close attention to controls. It requires clear data ownership. It requires architecture that supports end-to-end visibility. And it requires governance that can move from case-by-case intervention towards transparent rules wherever the risk permits.
The objective is not less governance.
It is governance capable of operating at digital speed.
The enterprise boundary is becoming part of the architecture
The State of Agility research becomes even more significant when it moves beyond internal functions.
Ninety-three per cent of respondents believe reducing fragmentation with strategic customers and suppliers is important or very important. The areas in which they see the greatest value from cross-company collaboration include innovation, go-to-market activity, R&D and sales and marketing.
For technology leaders, this extends the architecture problem beyond the enterprise perimeter.
Modern value creation increasingly depends on networks of suppliers, platforms, partners and customers. Data, workflows and decisions move across company boundaries. AI systems will increasingly operate across those same networks.
Yet the report shows that transformation remains overwhelmingly internal.
Seventy-eight per cent of respondents report being at some stage of adopting a new operating model to scale AI-driven value. 31.11% are conducting first experiments with cross-functional practices and AI solutions. 22.22% report successful pilots and 15.56% have initiated functional transformation. Only 7.41% report transformation coordinated at company level, and just 1.48% say transformation includes partners across the value stream.
The ecosystem ambition is therefore far ahead of ecosystem architecture.
That gap matters because the next generation of enterprise technology will not operate neatly inside organisational boundaries.
Supplier data, shared planning, joint innovation, commercial platforms, external AI services and connected customer ecosystems all create new questions around interoperability, security, trust, data rights, governance and accountability.
The technology function will increasingly be asked not only to secure and integrate the enterprise, but to enable controlled participation across an ecosystem.
This demands a more nuanced approach than simply opening access or building APIs. It requires a clear understanding of which information should move, which decisions can be shared, where accountability remains internal and how governance operates when multiple organisations participate in the same value stream.
Architecture is therefore becoming a mechanism for commercial coordination.
The CIO mandate is becoming an operating-model mandate
The hardest part of this transformation is unlikely to be technical.
The leading barriers identified in the study are organisational resistance to change, lack of skills and experience with agile methods, and organisational culture being at odds with agile values. Insufficient training, inadequate management support and governance constraints also feature prominently.
That matters because many technology transformations are still structured as implementation programmes when they are really behaviour-change programmes with technology inside them.
A successful AI platform can fail to scale if managers do not trust delegated decision-making. A self-service model can fail if functions refuse to surrender control. A shared data environment can fail if teams continue to protect local ownership. Automation can stall if policies are not redesigned. Cross-functional platforms can underperform if incentives remain functional.
For technology leaders, the mandate is therefore expanding.
CIOs and CTOs are increasingly being asked to design not only systems, but the conditions under which systems can create value.
That means working with leadership on decision rights, with finance on controls, with legal and risk on policy, with procurement on supplier ecosystems, with operations on workflows and with HR on skills and adoption.
It means identifying where the organisation is asking technology to compensate for structural problems that should be addressed directly.
It also means resisting a seductive but dangerous assumption: that because AI can perform more work, the enterprise automatically becomes more intelligent.
Intelligence at system level depends on whether insight can move into action.
The organisations best positioned to capture AI-driven value will be those that connect architecture, data, governance and operating-model design rather than treating them as separate transformation agendas.
Ledger Series Media saw the same themes emerge while supporting the State of Agility campaign. Direct engagement across senior technology, procurement, supply chain, transformation and leadership audiences repeatedly moved beyond AI itself towards governance, decision rights, silos and cross-functional execution. Nearly half of the completed survey responses recorded at that stage came through Ledger Series Media’s own campaign activity, giving the Group a substantial direct view into how senior professionals were approaching these challenges.
That experience reinforced why the State of Agility conversation matters across the Ledger portfolio. The questions are not confined to procurement. They concern how modern organisations convert increasingly powerful technology into coordinated business performance.
For Technology Ledger, that is the central implication of the 2026 findings.
The next technology transformation will not be defined only by better models, more automation, stronger data platforms or deeper integration. Those capabilities will be part of it. The defining shift will be whether technology leaders can help create an operating environment in which those capabilities translate into faster decisions, shared visibility, proportionate governance and coordinated value creation.
That is also why Ledger Series Media Group is supporting the discussion across its portfolio. Through Ledger Series Intelligence, the Group will continue building on themes raised by the research through market insight, executive commentary and practical perspectives on the operating questions organisations are now confronting.
As Alexander Barron notes in the report foreword, technological progress is moving quickly, but lasting transformation continues to depend on people, leadership, collaboration and the ability to adapt.
For technology leaders, that may be the defining challenge of the next phase.
AI can accelerate intelligence.
The CIO opportunity is to redesign the system that turns intelligence into action.
Explore the full 2026 State of Agility in Procurement & Supply
The report examines business agility, operating-model redesign, AI adoption, organisational fragmentation, supplier ecosystems, commercial collaboration and the practices organisations are using to create faster, more adaptive commercial systems.
