The SAITO Blueprint – A Practical Path Forward For AI Transformation

Stephen Ram
4 days ago
10 min read

From Pilots To Strategic AI Transformation

AI Technology Is Ready - But Organisational Capability Is Not

Every day, leaders are bombarded with noise - headlines about trillion-dollar AI opportunities, predictions of disruption, and endless technical detail. Everyone agrees AI has vast potential. But the truth is, most organisations still don’t know how to get from experimentation to ROI.

Across industries, CEOs, CFOs, COOs, CIOs, CROs, and transformation heads all describe the same reality: AI costs are rising, learning is high, but measurable business return is missing. Pilots are everywhere, but few deliver sustained impact - research from MIT, BCG, and Gartner shows around 85% fail to deliver ROI, and more than half are expected to be abandoned.

The issue isn’t the technology - it’s ready. The gap is the missing organisational capability to re-engineer the business to take full advantage of the significant opportunities AI enables.

Every leader is asking the same question: How do we move beyond technical pilots to achieve meaningful business results?

A Practical Blueprint- Strategic AI Business Transformation

After more than thirty years leading ROI-driven transformation - from business model redesign to enterprise IT - I’ve seen both the failures and the solutions. The pattern is clear- technology advances quickly, but strategic success depends on the organisation’s ability to adapt and build the new capabilities required to take advantage of it - capabilities often very different from those that exist today.

In conversations with senior leaders across sectors, the same point keeps surfacing the Central Strategic AI Transformation Function is missing.

This Blueprint has been created as a recommended approach that fills the gap - providing a practical framework leaders can implement. It’s about giving leaders a solution - not adding to the noise and the hype.

AI is a Fundamental Step Change

Throughout history, there have been many waves of innovation. Some - like the Industrial Revolution and the introduction of assembly lines - reshaped entire industries. Others, such as digital and data, have been more incremental. AI, however, sits firmly among the most transformative, offering a leap in scale comparable to those earlier industrial shifts.

AI is not just another technology wave. It represents a fundamental change in how businesses operate - from deterministic systems that follow rules to adaptive networks of AI agents that learn, optimise, and make decisions in real time.

But while opportunity has grown, organisations’ ability to capture it has declined. Early revolutions achieved near-total transformation; recent ones have seen success rates fall to around 30%. With AI, that figure has dropped again - to roughly 15%. The result is a realisation gap that’s wider than ever.

The AI technology is ready. What’s missing is the organisational capability to take advantage of it - to turn potential into performance. Closing that gap demands deliberate, decisive action, or the biggest opportunity in decades will remain unrealised.

Note: Values are conceptual indices (0-100) representing relative theoretical vs realised business benefit. Adoption windows are approximate and for illustration only. Empirical proxies may be drawn from: 1) historical productivity and TFP growth data (Allen, Crafts, OECD, Brynjolfsson & Hitt): 2) firm-level transformation success studies (McKinsay, BCG, Gartner); 3) AI productivity J-curve research (MIT, OECD).

Entering The Trough Of Disillusionment

After months of hype and inflated expectations, AI has reached an inflection point. The market is saturated with bold claims and rising costs, but the promised returns are proving harder to find. For many AI is adding to the cost base rather than transforming the business model.

AI is now entering the Trough of Disillusionment - the phase on the wider technology hype cycle where reality catches up with expectation. The hype cycle describes how every major innovation moves from early excitement and inflated expectations through a period of disappointment before stabilising into sustained value creation.

This isn’t failure; it’s a natural part of the innovation cycle - but it’s also the moment where the real work begins.

This trough happens at the start of every major innovation wave. Organisations do what’s easy - running pilots, proofs of concept, and experiments in innovation labs or IT teams. It feels productive, it’s fast to fund, and it creates visible activity. But it rarely moves the organisation any closer to real strategic transformation. When the investment mounts and business impact doesn’t, disillusionment follows.

The hype cycle doesn’t happen because expectations are too high, but because organisations assume early experiments will somehow deliver the step-change the business needs. The true transformation required to become an AI-led organisation - with all the benefits that brings - doesn’t happen organically. Rolling out tools and expecting revolution keeps you in the game, but it won’t change it. The path forward demands a defined destination: a clear strategy for how the AI-led enterprise will operate in the future, and deliberate action to start moving toward it.

Avoiding The AI Adoption Gap

A Likely Reduction In Adoption

The result of the “Trough of Disillusionment” phase is often a reduction in adoption. Gartner forecasts that around 60% of AI pilots will be shut down over the coming year. That doesn’t mean AI has failed - it means organisations have reached the limits of experimentation. Most will pause to regroup before moving forward again, creating a short-term dip in adoption as they recognise the need for a more strategic, enterprise-level approach. It’s the same inflection point seen in every major technology era: adoption falls before it rises.

Why A Strategic Approach Can Avoid it

This slowdown is avoidable. A few organisations have already shown how to break the pattern.

DBS Bank faced this same turning point during the digital era. Instead of scaling pilots, it rebuilt how the business worked - redesigning data, processes, and governance around measurable outcomes. The result was a step change in speed, quality, and ROI that redefined digital transformation in its industry.

Siemens has done the same with AI. By creating a central AI Accelerator to coordinate global projects, it replaced scattered pilots with a unified capability built on shared data, tooling, and measurement. The result: consistent, measurable impact - from predictive maintenance to energy optimisation.

The lesson is clear. Organisations that build strategic capability can bypass the trough altogether, moving directly from experimentation to unlock the benefits the new capability brings.

Where Are You On Your AI Maturity Journey?

Most organisations overestimate their AI maturity. Pilots and experiments give the impression of progress, but they’re usually isolated and tactical. Real maturity means knowing where you stand against the ultimate destination - becoming a genuinely AI-led organisation.

That’s a structural shift: open-heart surgery on the business, with new ways of working and redesigned processes. The real question is whether you’re moving forward on that path. Most organisations stall between stage two and stage three, where experimentation must give way to something strategic.

Every organisation needs to take a hard look at where it is today and ask: Do we have a defined path to the AI-led Enterprise end-state - or are we stuck in the experimentation phase?

This table outlines the five stages of AI maturity - from early concepts to full enterprise orchestration - helping you assess where your organisation realistically sits on the path to becoming AI-enabled.

The Strategic AI Transformation Office (SAITO)

AI transformation won’t just happen on its own. It requires a specialist function designed to manage the unique characteristics and challenges of AI - a technology capability that is still emerging, fast-moving, and deeply interwoven with business operations.

The Strategic AI Transformation Office (SAITO) is designed to provide that function, ensuring the organisation has the capabilities needed to turn ambition into measurable business performance. It bridges the gap between innovation and execution, aligning strategy, technology, data, and delivery into one coherent model.

The Blueprint is designed as a practical framework for establishing SAITO and equipping it to succeed.

The next section describes SAITO in more detail:

How Does SAITO Work? – The operating model that connects strategy, delivery, and governance through an eight-stage lifecycle.
Who Leads SAITO & What Is Their Skill Profile? – The cross-functional leadership structure.
What Is The SAITO Lifecycle? – The recurring stages of continuous improvement.
How Is SAITO Different To Traditional Transformation? – Ten additional focus areas specific to AI-driven change.
Where Does SAITO Sit & How Do We Stand It Up? – linking executive governance to operational execution and the practical steps to establish it quickly using existing resources.
What Does The SAITO Team Deliver? – Builds the capabilities to transition to an AI-led enterprise.

How Does SAITO Work?

AI transformation moves too fast for traditional corporate governance. New data, models, and risks appear daily. Strategy, design and operations evolve together. The only way to manage it is through a live, adaptive system.

At the centre of that system sits the AI Transformation Cycle - eight connected sequential capabilities that run in iterative cycles. Together, they form a repeatable rhythm of learning, direction, design, delivery, and control - helping leaders anticipate, visualise, and manage AI as a living, learning system for growth.

Who Leads SAITO & What Is Their Skill Profile?

Most organisations respond to transformation by forming a committee of senior representatives from each department, meeting every few weeks. That won’t work here. AI moves faster and cuts deeper. It demands a dedicated, full-time team with the skill, mandate, and authority to act across functions - not review from the sidelines.

SAITO is led by a small, cross-functional group of experienced innovators drawn from across the enterprise. They act on behalf of the executive, defining strategy and connecting it directly to delivery on the ground.

This isn’t a committee - it’s an operating engine for AI transformation. The team builds capability, drives progress, and keeps strategy, benefits, and delivery aligned in real time. It engages the executive on the major decisions that shape direction, ensuring momentum is maintained and learning is shared enterprise-wide. The following table shows the specific SAITO roles needed.

What Is The SAITO Governance Lifecycle?

SAITO runs on a multi-layered leadership governance rhythm - annual, quarterly, monthly, and sprint-by-sprint. Each layer reinforces the others, ensuring strategy, execution and decision-making stay aligned and continuously adaptive.

How Is SAITO Different To Traditional Transformation?

Moving from pilots to strategic AI isn’t about just launching a big programme. The issue is that most organisations try to run AI transformation using the same change methods built for traditional technology. But AI brings a new set of demands: it moves faster, goes deeper into the operating model, needs measurable ROI and carries far higher cultural sensitivity.

Each of these focus areas creates exposure that traditional approaches aren’t built to manage. Any one done badly can derail progress entirely. These ten capabilities define the areas every organisation must strengthen to deliver AI transformation safely and successfully.

Where Does SAITO Sit & How Do We Stand It Up?

Where SAITO Needs to Be positioned

SAITO must sit close to the executive - reporting directly to the CEO, COO, or equivalent. It spans all major internal functions - data, delivery, operations, risk, finance, HR, and technology - as a single, cross-organisational function for AI-led transformation. It cannot sit in IT or an innovation centre. This is open-heart surgery on how the business runs. The role of SAITO is to keep strategy and execution continuously connected - ensuring the executive knows what’s really happening, and delivery teams focus on what matters most.

How Long Does It Take To Stand SAITO Up?

Standing up SAITO isn’t a long process. It’s about deciding to do it and making it happen, assembling the right people and following the blueprint. An organisation can move from decision to operation in six to eight weeks.

What Does The SAITO Team Deliver?

SAITO’s engine room is the Knowledge Centre. You don’t magically develop the ability to harness a new technology; you build it deliberately and systematically.

The Knowledge Centre provides the structure for that build - the deliverables, artefacts, and methods that turn ambition into capability. Each deliverable aligns to one or more of the ten critical capabilities that differ from traditional transformation, ensuring they’re designed, tested, and embedded.

The deliverable structure is tool-agnostic and organisation-neutral, integrating with any structure or technology stack. The framework applies consistently across business units and governance models, forming the foundation for scalable AI delivery. The table below lists the deliverables that bring these capabilities to life across SAITO’s eight building blocks.

Closing Thoughts

Blueprint Summary – A Practical Path Forward

Six months on, the structure is in place. The Strategic AI Transformation Office is operating as the hub of direction, measurement, and delivery. AI is no longer scattered across the organisation - it’s connected to strategy, investment, and outcomes. Leaders can see progress clearly, make decisions quickly, and focus resources where they have the greatest impact. The pace feels steady and purposeful. Governance is lighter but more effective, teams are aligned, and value is visible. The organisation has moved from pilots and proofs of concept to a system that learns and improves continuously.

This is what strategic capability looks like in practice. It replaces enthusiasm with evidence, pilots with performance, and activity with alignment. The first step is recognition. The next is action. Those who take it now will move faster, learn faster, and deliver results that last.

A Personal Note

On a personal note, this is a space I’ve been passionate about for decades. For over 30 years I’ve worked in ROI-led change and transformation, building the structures, methods, and predictive intelligence needed to link strategy, delivery, and measurable outcomes.

Ten years ago, I built a business around that belief. Since then, we’ve applied our ROI-led approaches with leading banks, insurers, and big-four consultancies, helping them deliver transformation through realistic costs, measured ROI, and controlled risk.

My offer is simple - My personal goal is to help organisations take advantage of what I see as the most significant step-change in business performance in generations. If you’d like to explore this in a bit more detail, or ask questions, click below and we’ll arrange a short follow-up - or feel free to reach out to me directly on email or LinkedIn (details below).

If you need more support to get started Cedus can provide the Blueprint, Knowledge Centre, capability frameworks, tools, and templates that underpin this approach. But this isn’t a sales document; the Blueprint stands on its own and is designed to be used.

Read this article offline as a pdf: https://tinyurl.com/5hf3urj9

About the Author

Stephen Ram

Co-Founder of Cedus, is a veteran change and transformation leader with 30+ years of experience driving measurable ROI across global organisations. After witnessing how traditional change efforts focus on activity over value, he co-founded Cedus to put ROI at the centre of transformation. Today, Stephen helps leaders gain predictive clarity, make faster decisions, and turn change into a true competitive advantage.