The Architecture Marketplace:

How to Build Cathedrals, Harness Hares, and Run a Thriving Bazaar.

This long-form Insight brings together a series of concise articles providing experienced based views on how to run smarter, not just faster to deliver sustainable transformation.

On this page

Introduction
Speed Without Anchors
Complexity or Progress?
Innovation or Distraction?
ROI Illusions
Governance That Moves
Culture ‘v’ Transformation
Mind the Skills Gap
Beyond the Sales Pitch
Escaping the Legacy Trap
Breaking Change Fatigue
What Gets Measured Gets Done
Continual Transformation
The Strategy and Delivery Gap
Over-Promise, Under-Deliver
Lone Wolf Transformation Fails

Introduction

Over the last three decades I’ve been involved in transformation programmes of every flavour — from government mega-projects to high-growth scale-ups, from banks and telcos to tech startups. Different industries, different budgets, different cultures. But one pattern shows up again and again: most programmes fail not because of technology, but because they lean too heavily on one way of thinking.

Some leaders build Cathedrals.

They produce majestic plans, beautifully detailed blueprints, and multi-year delivery schedules. Everything is neat on paper. Every milestone is defined. Every dependency tracked. It looks impressive — until reality intervenes. By the time the Cathedral is built, the market has moved on, the customers have changed, and the original design is already out of date.

Others chase Hares.

Quick wins. Fast sprints. Something to demo next month. These efforts generate energy and enthusiasm, and they can be brilliant for building momentum. But too often they fizzle out. The hare burns bright but rarely sustains. What’s left behind is fragmented, hard to scale, and not connected to a bigger picture.

And then there’s the Bazaar.

Messy, noisy, full of energy and innovation. People collaborating, testing, arguing, adapting. New ideas pop up daily. Some work, some don’t. The Bazaar thrives on diversity and experimentation. But without any structure, it quickly becomes chaos. Good ideas are lost. Duplicated effort drains resources. Decision-makers get overwhelmed.

The brutal truth is this: none of these models work in isolation.

Too much Cathedral, and you get rigidity and irrelevance.
Too much Hare, and you get shallow wins without endurance.
Too much Bazaar, and you get noise without direction.

The organisations that succeed treat transformation as a marketplace of ideas.

Think of it like this:

The Cathedral provides structure — principles, guardrails, long-term clarity.
The Hares provide speed — energy, experimentation, proof points.
The Bazaar provides adaptability — inclusion, creativity, a space for ideas to compete.

When you bring those elements together, you don’t get a project plan, you get an ecosystem. One where vision and execution reinforce each other. One where people are encouraged to innovate, but always within a framework that connects back to core business value.

This is also where Data and AI become critical. Left unchecked, AI pilots can create exactly the kind of fragmentation and chaos that sink transformations. But inside a marketplace model, they become accelerants:

Pilots can be run quickly, tested against business capabilities, and either scaled or retired.
Data can reveal which experiments have real potential and which are smoke and mirrors.
AI analytics can provide early signals — spotting bottlenecks, surfacing ROI illusions, predicting delivery risks.

The difference is that the organisation has built a place for these things to plug in — rather than bolt-ons that eventually break off.

So how do you build an architecture marketplace in practice? A few patterns I’ve seen work:

Define principles early. Treat them as living guardrails, not fixed rules. Principles give people confidence that they’re aligned, even when innovating at speed.
Enable parallel delivery. Instead of a single critical path, create capability boundaries where teams can run in parallel without tripping over each other.
Foster a culture of experimentation. Encourage new ideas, but insist they’re tested against business outcomes.
Use data and AI as filters. Not everything should scale. The discipline is knowing what to stop as much as what to grow.

Done well, the marketplace becomes a self-reinforcing loop: vision creates direction, quick wins create momentum, and innovation creates resilience.

The real takeaway? Architecture must be a platform, not a prison.
If you design transformation like a Cathedral, you’ll be too slow. If you chase only Hares, you’ll burn out. If you only run a Bazaar, you’ll drown in noise. But if you create a marketplace where all three coexist, you build something both visionary and adaptable.

And in a world where business models shift faster than ever, that balance isn’t just useful — it’s survival.

Speed Without Anchors

I’ve lost count of the number of transformation programmes I’ve worked on and witnessed that look great on the surface — with slick demos, fast-moving delivery teams, confident slide decks — but stall and sometimes collapse as soon as they’re meant to scale.

The problem? They confuse speed with progress.

Speed is intoxicating. Executives love it. Boards love it. Shareholders love it. Who wouldn’t want to see something delivered in weeks instead of months, or in months instead of years? The trouble is, when transformation runs too fast without a foundation, all you’re doing is hitting the wrong target quicker.

Let’s call this the “speed without anchors” trap.

Why It Happens

It happens because delivery teams are under enormous pressure to show results early. No one wants to wait. So shortcuts are taken:

Capability models aren’t consulted.
Architecture isn’t engaged.
Minimum Viable Product is redefined as “ship something now, figure out the rest later.”

On paper it looks like progress. In reality, it creates fragile solutions that can’t scale, don’t integrate, and don’t deliver lasting value.

It’s like building a speedboat without checking the map. You’re moving fast, but you might be heading in the wrong direction entirely.

The AI and Data Dimension

Nowhere is this trap clearer than in Data and AI. I’ve seen AI teams spin up amazing proofs of concept in a matter of weeks. The demos are jaw-dropping. But they’re built on isolated datasets, disconnected from the core enterprise architecture.

The result? Clever science experiments. Nothing more. When the time comes to deploy into production, integrate with real data, or connect to business processes, the entire thing collapses.

AI is not a magic wand. If it’s not anchored to your core capabilities and data platforms, it becomes yet another shiny object that drains investment without delivering return.

What Works Instead

Anchoring doesn’t mean slowing down. Done right, anchoring actually accelerates value because you’re building for scale from day one. The programmes I’ve seen succeed do a few things differently:

Tie everything to a capability map. If you can’t trace a piece of delivery back to a business capability, it’s probably waste.
Use architecture reviews as accelerators, not brakes. Architects shouldn’t be the people who say “no.” They should be the people who ask, “How do we make this scale?”
Redefine MVP. Minimum viable product doesn’t mean “throwaway prototype.” It means “a foundation that works today and can grow tomorrow.”
Connect AI and data pilots to the core from the start. Don’t let data science happen in a silo. Build integration points into your operating model.

Anchors Aren’t Shackles

One of the myths about anchoring is that it slows you down. In fact, it’s the opposite. Anchors give you direction. They stop you wasting energy on things that won’t last. They mean that when you do move fast, you’re moving fast in the right direction.

The best analogy I’ve found is sailing. A boat without an anchor drifts. A boat with a well-set anchor can hold steady in storms and reposition with intent. Transformation needs the same.

The Takeaway

The brutal truth is this: speed without anchors creates waste. It looks good for a quarter or two, but it leaves nothing sustainable behind.

If you want transformation to stick:

Map before you move.
Build for scale, not just for demo.
Treat AI as part of the core, not as a side project.

Speed is not the enemy. But speed without direction, without anchoring, is just noise. The organisations that master this know how to move fast and hold course. They don’t just sprint — they sprint on the right track.

And that’s what separates transformation that delivers a press release from transformation that delivers lasting business value.

Complexity or Progress?

Transformation programmes love complexity.

The more layers, the more detail, the more governance checkpoints, the more impressive it all looks. But here’s the hard truth I’ve learned over decades in this field: complexity is often a disguise for a lack of real progress.

I’ve sat in rooms where architects proudly walk through 200-page slide decks, every dependency mapped, every interface named, every risk logged. It feels thorough. It feels serious. But when you look closely, months have passed, budgets are burning, and very little tangible value has actually been delivered.

Complexity creates the illusion of progress. Real progress is much simpler.

Why It Happens

There are a few recurring reasons why transformation projects fall into the complexity trap:

Detail is mistaken for clarity. Teams assume that more diagrams, more artefacts, and more committees will make things clearer. Often it just adds noise.
Stakeholders want to justify their involvement. Every department adds their own layer of requirements, controls, and reporting — not because it’s needed, but because they don’t want to be left out.
Integration is handled reactively. Instead of designing clean handshakes and automation up front, dependencies are patched together in flight, which breeds ever more layers of coordination.

The end result is a machine that looks sophisticated but moves painfully slowly.

The AI and Data Dimension

Data and AI make this worse and better, depending on how you use them.

Worse, because AI pilots can become yet another silo. I’ve seen organisations set up separate governance boards just for “AI initiatives” — creating duplication, confusion, and more meetings without addressing the core integration problem.

Better, because AI (used properly) can actually cut through complexity. Predictive analytics can highlight process bottlenecks before they become crises. Process mining can show which steps add value and which are pure waste. Machine learning can forecast which initiatives are most likely to fail, saving time and money.

But that only works if the data is clean and the enterprise has the discipline to listen to the signals rather than layering on yet more process.

What Works Instead

The organisations that escape the complexity trap do three things differently:

Strip governance back to essentials. Focus on the few principles that matter — capability alignment, security, scalability — and let the rest flow.
Automate wherever complexity is unavoidable. If you must have 20 handoffs, fine — but don’t make them manual. Use integration and automation to take the friction out.
Prioritise value-based sequencing. Deliver the integrations that unlock the most business value first, even if they aren’t the neatest from a design perspective.

Complexity can’t be eliminated altogether, but it can be managed. The key is to distinguish between necessary complexity (which adds resilience) and performative complexity (which adds nothing).

Anchoring Simplicity

Here’s the pattern I’ve seen: successful transformations tend to get simpler as they go. Early chaos is inevitable, but over time they converge on a small number of guiding artefacts that everyone uses — capability models, a single architecture framework, a shared roadmap.

Failing transformations go in the opposite direction. They start with a few documents, then grow to hundreds. More committees, more dashboards, more checklists. Eventually the machine becomes so heavy it collapses under its own weight.

Simplicity is not about dumbing down. It’s about focusing energy where it matters most.

The Takeaway

If you feel like your transformation is “stuck,” ask yourself: are we actually moving, or are we just generating more complexity?

Because:

More meetings ≠ more progress.
More artefacts ≠ more alignment.
More governance ≠ more control.

The transformations that deliver value are the ones that learn to simplify. They use data and AI to highlight waste, they automate the unavoidable, and they keep governance sharp and light.

Everything else? That’s noise masquerading as progress.

And the sooner you strip it back, the faster you’ll move toward outcomes that actually matter.

Innovation or Distraction?

Every transformation programme I’ve been part of has, at some point, been tempted by the “next big thing.”

Blockchain. Metaverse. Digital twins. Generative AI. The names change, but the pattern doesn’t. Leaders chase shiny objects, convinced that this new technology will solve deep structural problems.

The trouble is, shiny objects are seductive but rarely strategic. They distract from the hard work of aligning strategy, culture, and execution. They scatter focus and fragment investment. And nine times out of ten, they deliver very little in return.

Why It Happens

The psychology is simple:

Leaders want to be seen as innovative. Backing the latest tech trend signals vision and modernity — even if the business case is thin.
Vendors push hard on hype cycles. When the big consultancies and analysts say, “This is the future,” it’s tough for boards to resist.
There’s no consistent framework for evaluation. Without a disciplined way to assess whether new tech aligns with core capabilities, decisions get made on gut feel and fear of missing out.

The result is a portfolio of disconnected pilots, half-finished initiatives, and business cases that never materialise.

The AI and Data Dimension

AI is the ultimate shiny object right now. Everyone feels they need an AI strategy — yesterday. I’ve watched companies spin up AI projects simply because “the board asked for it.”

The outcome? Expensive science experiments. A chatbot that nobody uses. A predictive model that isn’t connected to operations. A dashboard that looks impressive but doesn’t drive decisions.

The irony is that AI has enormous potential — but only if it’s tied directly to core business capabilities and real problems. If it’s treated as an isolated “innovation initiative,” it just becomes another distraction.

What Works Instead

The organisations that get this right filter every new idea through three tough questions:

Does it fit our strategy? If a shiny object doesn’t align with long-term business outcomes, park it.
Does it enhance our core capabilities? If it doesn’t strengthen something we already do or enable something we must do, it’s probably noise.
Can we measure value fast? If the business case is all “potential” and no “proof,” run a controlled pilot with clear kill criteria.

The principle is simple: explore broadly, but scale selectively.

Guardrails Without Killing Curiosity

I want to be clear: innovation is essential. No organisation survives without testing new ideas. The problem isn’t innovation itself — it’s undisciplined innovation.

The best transformation leaders I’ve worked with know how to build guardrails that encourage experimentation but prevent chaos. They create what I call an “architecture marketplace”: a space where new ideas can be tested quickly, connected to core systems early, and scaled up if they deliver value.

That’s how you separate genuine breakthroughs from shiny distractions.

A Story from the Field

A few years ago, I worked with a financial services firm that got caught up in blockchain hype. They launched three separate pilots, all expensive, none aligned to customer needs. Millions were spent. Not a single pilot scaled.

Meanwhile, their real problem — fractured customer data across dozens of systems — went unaddressed. That was the thing driving customer churn and operational inefficiency. But because blockchain was exciting and sexy, leadership attention drifted.

Contrast that with a retailer I supported more recently. They were curious about AI, but instead of launching random pilots, they asked: “Where does AI make us faster or more accurate against our strategy?” The answer was inventory forecasting. They ran a small pilot, proved it worked, and then scaled it across the business. Within a year, stockouts dropped by 25%.

Same technology. Very different outcomes.

The Takeaway

Shiny objects aren’t going away. There will always be a new technology trend that promises to change everything. The job of a transformation leader isn’t to ignore them — it’s to test them with discipline.

Filter against strategy and capabilities.
Pilot fast, kill faster.
Demand evidence of business value before scaling.

Because the truth is, innovation without alignment is distraction. But innovation with discipline? That’s where transformation becomes real.

ROI Illusions

Every transformation programme starts with a business case. The numbers look compelling: reduced costs, new revenue streams, happier customers, higher NPS, faster processes. On paper, the ROI is undeniable.

But walk into most boardrooms two or three years later, and you’ll hear the same quiet frustration: “Where are the benefits we were promised?”

The milestones were ticked off. The technology was delivered. The consultants were paid. Yet the bottom line looks strangely unchanged. The ROI never shows up.

This isn’t bad luck. It’s a recurring pattern — one I’ve seen play out across industries for decades.

Why It Happens

There are some common traps:

Optimism bias in the business case. Benefits are inflated to win funding. Costs are underestimated to make the numbers work. Everyone knows it, but the cycle repeats.
Focus on delivery, not outcomes. Teams celebrate go-lives, not whether customers are happier or costs are actually down.
Benefits ownership is vague. No single leader is accountable for turning promised outcomes into reality — so nobody does.
Tracking stops too early. Benefits measurement ends when the project closes, just when it should be starting.

The result is what I call ROI illusions — benefits that exist on slides but never land in the business.

The AI and Data Dimension

AI and data should be the answer here — but often they become part of the illusion.

I’ve seen dashboards full of KPIs that look impressive but have no link to financial reality. I’ve seen AI models used to forecast benefits, but the assumptions are so optimistic they become self-fulfilling nonsense.

That said, when applied with discipline, data and AI can expose the illusion early. Predictive analytics can flag when expected cost savings aren’t materialising. Real-time dashboards can track whether customer sentiment is actually improving. Machine learning can compare expected vs actual ROI across initiatives to identify which patterns consistently under-deliver.

The technology exists to keep ROI honest. The question is whether leaders are willing to face the truth.

What Works Instead

From experience, the organisations that consistently realise benefits treat ROI as a living, breathing commitment — not a static promise.

Assign clear ownership. Every promised benefit has a name next to it. Not a department. A person. Someone accountable.
Track continuously, not just at closure. Measurement continues for months, even years, after delivery.
Re-baseline when reality changes. If assumptions turn out to be wrong — and they often do — update the plan instead of pretending.
Tie AI and data directly to business KPIs. Not vanity metrics. Not activity measures. Real customer, financial, and operational outcomes.

A Story from the Field

I once worked with a large public-sector organisation that launched a multi-year transformation. The business case promised £200m in savings. Three years in, they were delivering only about 30% of that.

The problem wasn’t technology — the systems worked fine. It was that the benefits were based on heroic assumptions: people would adopt new processes instantly, customers would change behaviour overnight, and costs would drop without resistance. None of it held true.

Contrast that with a private healthcare provider I supported more recently. Their transformation case was conservative. Every benefit had a named owner. AI-powered dashboards tracked savings and customer NPS weekly. When adoption lagged, they adjusted. Within 18 months, they had exceeded their ROI targets — not because the tech was better, but because the discipline was real.

The Takeaway

If you’re leading a transformation, ask yourself: are your benefits real, or are they illusions?

Because:

Benefits don’t deliver themselves.
ROI isn’t static — it needs constant management.
Data and AI can keep you honest, but only if you’re willing to listen.

The most successful transformations are not the ones with the most ambitious business cases. They’re the ones with the most disciplined benefit tracking and the courage to admit when assumptions don’t hold.

That’s how you turn ROI from a slide-deck illusion into measurable, bankable reality.

Governance That Moves

Governance. It is just one of those words that makes people groan.

It conjures images of endless committees, red tape, sign-off gates, and progress slowed to a crawl. Yet without governance, transformation descends into chaos: duplicated effort, fragmented systems, uncontrolled risk.

The challenge isn’t governance itself. The challenge is bad governance — governance designed for control in a static world, not for speed in a dynamic one.

If transformation is to succeed, governance has to move. It has to guide without suffocating. It has to protect without paralysing.

Why Governance Fails in Transformation

Across industries, I see the same patterns:

Static rules in a fluid environment. Governance is written once, applied rigidly, and never adapted to new realities.
Centralised bottlenecks. Decisions pile up at the top. Senior leaders become the only ones “trusted” to approve, so everything slows.
Focus on compliance over outcomes. Teams spend more time proving they followed the process than proving they delivered value.

The result is predictable: delivery loses pace, innovation stalls, and the business starts to bypass governance entirely just to get things done.

The AI and Data Dimension

Governance is often painted as paperwork. But modern governance can — and should — be powered by data.

Imagine risk management that isn’t based on quarterly reports but on real-time analytics. Imagine compliance gaps flagged automatically by AI, not discovered in audits 12 months later. Imagine decision-making dashboards that give leaders a live view of where projects stand, risks they carry, and value they’re creating.

This is what I call governance that moves. It’s lighter, faster, and smarter — not because the principles are looser, but because data and AI make the system more responsive.

What Works Instead

To keep governance alive and relevant in transformation, leaders need to shift from control-centric to principle-driven approaches:

Guardrails, not handcuffs. Define principles (e.g., “every solution must integrate to the data platform”) rather than detailed step-by-step approvals.
Distribute decision rights. Push decisions as close to the work as possible. Empower teams to make choices within agreed boundaries.
Use AI to amplify visibility, not paperwork. Automate risk detection, compliance checks, and dependency mapping so governance is based on evidence, not opinion.
Adapt continuously. Governance frameworks should evolve with delivery pace and external context. What worked last year may not work now.

A Story from the Field

I once worked with a telecoms firm whose governance board became infamous. To get approval for a new initiative, a team had to submit 200+ pages of documentation. Reviews took months. By the time approval came through, market conditions had shifted. Projects launched already out of date.

Eventually, leaders realised the governance model itself was the problem. They reset. Approval moved from documents to principles. AI-driven dashboards gave executives visibility into risks in real time. Teams could start work immediately within agreed guardrails.

The result? Delivery speed increased by 40%. Risk exposure dropped because issues were flagged earlier. And, perhaps most importantly, people no longer saw governance as the enemy — they saw it as an enabler.

The Takeaway

Governance isn’t optional. In highly regulated industries, it’s a necessity. But the choice leaders have is this: will your governance be a brake, or will it be a steering wheel?

Rigid, static governance kills transformation.
Principle-based, adaptive governance accelerates it.
Data and AI make governance lighter, faster, and smarter.

Transformation needs direction, not delay. If governance doesn’t move with the business, the business will move without it.

Culture 'v' Transformation

There’s an old saying in business: “Culture eats strategy for breakfast.” In transformation, it’s even harsher: “Culture eats transformation for breakfast.”

The truth is, most transformations don’t fail because the technology is wrong, or because the strategy is weak. They fail because the culture resists, rejects, or quietly ignores the change.

Culture is either the multiplier of transformation or the killer of it. Which one it becomes depends on how leaders engage with it.

Why Culture Blocks Transformation

Here are the most common patterns I’ve seen:

Underestimating inertia. Leaders assume people will adopt new ways of working simply because they’re told to. They won’t.
Talking process, not people. Communications focus on milestones, methodologies, and governance — not how people’s day-to-day lives will change.
Invisible wins. Teams achieve progress, but nobody celebrates it. Change fatigue sets in because all people see is the grind.
Cultural friction. New operating models often clash with old mindsets: risk-averse vs agile, siloed vs collaborative, command-and-control vs empowered.

Left unaddressed, these forces drain momentum and sink transformation efforts — no matter how solid the architecture or compelling the strategy.

The AI and Data Dimension

AI adoption is particularly sensitive to culture. Teams can view it as a threat — automation coming to replace jobs, or algorithms making opaque decisions.

I’ve seen AI initiatives stall not because the models didn’t work, but because employees didn’t trust them. “The AI says so” is not enough to build confidence.

On the flip side, when AI is introduced transparently — showing how it augments people rather than replaces them — adoption accelerates. People see it as a tool, not a threat. Data and AI become cultural assets, not cultural flashpoints.

What Works Instead

Culture doesn’t change itself. Leaders need to make it part of the transformation design, not an afterthought. Here’s what works:

Engage influencers early. Every organisation has cultural leaders — not always the ones with titles. Bring them into shaping change from the start.
Celebrate quick wins. Make success visible. Show progress in ways people can feel, not just read about in reports.
Link to personal benefit. Don’t just say, “This will improve efficiency.” Show how it makes someone’s job easier, faster, or more rewarding.
Position AI as augmentation, not replacement. Explain clearly how AI supports human judgment and frees people from low-value work.
Model the new behaviours at the top. Culture cascades. If leaders don’t live the change, nobody else will.

A Story from the Field

At a large financial services company, I watched a multi-year transformation stall because employees saw it as “management’s project.” They didn’t feel part of it. Every communication was about process and milestones. No-one explained how it would actually make their lives better.

Resistance was subtle but deadly: passive disengagement. Teams did the minimum. Adoption lagged. Benefits never landed.

Contrast that with a healthcare provider I supported. Their leaders treated culture as a design principle. They ran workshops with frontline staff before finalising any major changes. They celebrated small wins weekly. AI was introduced not as a replacement for clinicians, but as an assistant — flagging anomalies, helping them focus on patients.

The cultural shift was palpable. People leaned in. They felt ownership. The transformation stuck because the culture carried it forward.

The Takeaway

Transformation is not just about systems and structures. It’s about people.

Culture can block or multiply transformation.
Ignoring culture is a guaranteed path to failure.
Feeding culture with engagement, visibility, and trust turns it into your biggest asset.

And when it comes to AI, culture is the deciding factor. Introduce it badly, and it will be resisted. Introduce it transparently, and it will be embraced.

The most successful transformations are those where leaders don’t just feed the strategy. They feed the culture.

Mind the Skills Gap

When transformations fail, leaders often point to strategy, funding, or technology. But more often than not, the real culprit is simpler — and far more predictable.

It’s the skills gap.

The gap between the capabilities an organisation has today and the capabilities it needs tomorrow. The gap between designing a new model and actually running it. The gap between ambition and ability.

And the truth is, this gap is rarely a surprise. It’s visible from the start. Yet too many leaders ignore it until it’s too late.

Why Skills Gaps Sink Transformations

I’ve seen the same mistakes across industries:

Skills assessment comes too late. Organisations map the new operating model but only then ask, “Do we have the people to run this?”
Training budgets are cut first. When projects run over, investment in people is sacrificed to hit deadlines.
Assumptions replace analysis. Leaders assume “people will pick it up” — underestimating the shift in mindset, tools, and behaviours required.
Data and AI are treated as niche. Teams are expected to “use AI” without understanding the data literacy and governance disciplines that underpin it.

The result? New systems are delivered, but adoption is weak. Innovation stalls. Transformation benefits evaporate because people simply don’t have the skills to make the new model work.

The AI and Data Dimension

This problem is amplified in the era of AI.

Successful AI adoption isn’t just about hiring data scientists. It requires a workforce fluent in data: business leaders who can ask the right questions, managers who understand how to interpret AI outputs, frontline teams who trust — and know when to challenge — algorithmic recommendations.

Without that broad base of literacy, AI becomes either a black box (“just trust it”) or a toy for a small specialist team. Neither delivers business value at scale.

Skills gaps don’t just slow AI adoption. They kill it.

What Works Instead

Closing the skills gap requires intent from day one of transformation, not as an afterthought. Here’s what works:

Run skills gap analysis up front. Treat it like capability mapping. Know what you have today and what you’ll need tomorrow.
Make training part of delivery, not optional. Every new capability should have an accompanying skills development plan.
Invest in data literacy as a baseline. Everyone — not just technical staff — should understand the basics of data, AI, and automation.
Build cross-functional teams. Pair business domain experts with technical specialists to accelerate adoption and knowledge transfer.
Reward learning. Transformation is disruptive. Recognise and reward people who lean into building new skills.

A Story from the Field

At a logistics company I advised, leadership invested heavily in a new AI-driven forecasting system. The technology worked brilliantly in the lab. But once deployed, adoption collapsed. Why? The operations team didn’t understand how to interpret the outputs. They didn’t trust the recommendations. They defaulted back to manual spreadsheets.

The issue wasn’t the technology. It was the skills gap.

Contrast that with a retailer who treated skills as a core pillar of their digital transformation. Before rolling out AI tools, they trained all managers in data literacy. They ran scenario-based workshops to build trust in algorithmic recommendations. They created career development paths into data-centric roles.

When the technology launched, adoption was immediate. Teams felt confident using it. The benefits landed quickly — not because the tech was better, but because the people were ready.

The Takeaway

Every transformation is a skills transformation. Technology changes are easy compared to the human capability shifts required.

Skills gaps are predictable — and avoidable.
AI and data require literacy across the organisation, not just expertise in pockets.
Transformation without people development is just an expensive experiment.

If leaders want transformation to succeed, they need to stop treating skills as an afterthought. The missing link isn’t strategy, funding, or technology. It’s the ability of people to actually make the change work.

Beyond the Sales Pitch

Walk into any major transformation programme and you’ll find the vendor presentations.
Slick demos. Bold promises. ROI slides with curves that only ever go up. It looks convincing. The technology seems ready-made to solve your problems. The sales pitch promises faster delivery, lower costs, and seamless integration.

But too often, it’s a mirage.

Because when the excitement of the demo fades and the contracts are signed, reality sets in: deployment takes longer, integration is messier, outcomes are harder to measure. Technology isn’t the problem — it’s the assumptions leaders made based on the pitch.

Why the Vendor Mirage Happens

I’ve seen the vendor mirage trap across multiple industries. The patterns are clear:

Over-reliance on demos. Leaders see a polished product demo and assume it will translate directly into their environment.
Feature-first contracts. Deals are signed on the basis of features, not outcomes — which means success is measured by deployment, not business value.
No internal capability. Organisations lack the skills to challenge vendor claims, so assumptions go untested.
Data reality ignored. AI platforms in particular are often pitched as plug-and-play, but without clean, connected data, the results fall short.

The vendor mirage isn’t about bad technology. It’s about buying into promises without building the internal muscle to validate, adapt, and own the transformation.

The AI and Data Dimension

Nowhere is the mirage stronger than with AI.

Vendors pitch AI as a silver bullet: predictive, intelligent, automated. But what they don’t always emphasise is that the magic only happens if your organisation has the right data, governance, and integration foundations.

Without them, AI becomes just another dashboard. The outputs look impressive, but the business impact is minimal.

AI doesn’t fail because the algorithms don’t work. It fails because the environment it lands in isn’t ready. And vendors rarely tell you that part of the story.

What Works Instead

Avoiding the vendor mirage isn’t about distrusting technology providers. It’s about balancing optimism with realism and building the internal ability to own the solution. Here’s what works:

Validate with independent pilots. Don’t take the demo at face value. Run small-scale tests in your environment with your data.
Contract on outcomes, not features. Tie agreements to measurable business results, not just the successful delivery of a system.
Build internal capability. Develop teams who can assess vendor claims, challenge assumptions, and adapt solutions to your context.
Make integration part of the plan. Don’t treat it as an afterthought. Design upfront how the new technology will connect to your architecture, data, and processes.
Treat vendors as partners, not saviours. They bring tools. You bring context. Transformation happens when both sides contribute.

A Story from the Field

At a large retailer, leadership invested millions in a “next-generation” customer analytics platform. The demo was flawless. The ROI case looked irresistible. But once deployed, the system struggled. Why? Their customer data was fragmented across silos. Integration took years. By the time the platform worked, the market had moved on.

The technology was solid. The failure was assuming it would work straight out of the box.

In contrast, a telco I worked with took a different approach. They ran a three-month pilot with real data before committing. They brought in external validation to test vendor claims. They built internal data governance capability before rollout.

When the platform launched, adoption was smooth. The promised benefits materialised because they had built the foundations to make it real.

The Takeaway

The vendor mirage is seductive. Demos dazzle. Promises reassure. But transformation is not bought off the shelf.

Technology is an enabler, not a solution in itself.
ROI doesn’t come from features, but from integration, adoption, and outcomes.
AI in particular depends on your data reality, not the vendor’s pitch.

The organisations that avoid the mirage are the ones that combine external tools with internal ownership. They validate, adapt, and build capability.

Because in the end, vendors can sell you technology. But only you can deliver transformation.

Escaping the Legacy Trap

Every organisation has legacy systems. Some are decades old, built on technologies no graduate has learned in years. They run quietly in the background, processing transactions, handling logistics, or powering customer interactions.

And here’s the uncomfortable truth: they usually work.

That’s the paradox of legacy. On the one hand, it’s stable, proven, and business-critical. On the other, it’s brittle, outdated, and seen as a barrier to change. Too often, leaders fall into the “legacy trap”: ripping out systems wholesale in the pursuit of modernisation, only to destabilise the very business they’re trying to transform.

Why the Legacy Trap Happens

I’ve seen transformations fall into this trap repeatedly. It usually happens because:

All-or-nothing thinking. Legacy is framed as an obstacle to be eliminated rather than an asset to be managed.
Underestimating dependencies. Leaders don’t always appreciate just how interconnected legacy systems are. Pull one thread, and ten more unravel.
Shiny-object distraction. New technologies are prioritised for their promise, without a clear plan for how they’ll integrate with what already works.
Fear of missing out. Executives want to be seen as innovative, so they rush to replace instead of evolve.

The result? Ballooning costs, disrupted operations, frustrated employees, and customers left in the lurch.

The AI and Data Dimension

The rise of AI only sharpens the legacy dilemma.

On the one hand, AI depends on data. Legacy systems often hold the richest, most complete data sets in the organisation. On the other, those systems are rarely designed to make data accessible in modern, structured, and governed ways.

This is where smart leaders flip the script. Instead of smashing legacy systems, they use AI and automation to extend their life. Tools can extract insights from old databases, wrap APIs around core systems, and automate manual processes that previously acted as bottlenecks.

AI doesn’t remove the need for modernisation. But it can buy time, reduce risk, and allow transformation to happen as evolution — not revolution.

What Works Instead

Escaping the legacy trap means shifting mindset. The goal isn’t to break what works. It’s to create a pathway where the old and new can coexist — until the transition is safe, cost-effective, and value-driven.

Here’s what works:

Map dependencies first. Understand what relies on your legacy systems before you touch them. No surprises.
Phase modernisation. Don’t rip and replace. Build bridges that allow legacy and modern systems to run in parallel.
Use AI to extend, not just replace. Automate data extraction, create real-time dashboards, and wrap APIs around legacy to make it more usable.
Focus on business value. Don’t modernise for the sake of modernisation. Prioritise areas that directly impact customers, cost, or speed.
Communicate the journey. Teams trust the plan more when they see legacy treated as an asset, not a problem.

A Story from the Field

At a financial services organisation, leadership attempted to replace a 25-year-old core banking platform in one go. The system was old but had near-perfect uptime. The replacement programme dragged on for years, costing hundreds of millions, with constant delays. Eventually, it was shelved — while the legacy system quietly kept running.

By contrast, a healthcare provider I worked with chose a different path. They didn’t try to rip out their legacy systems. Instead, they built a data integration layer that pulled information into a modern cloud environment. They used AI to surface insights from previously inaccessible records. Over time, they gradually migrated functions — but only when it made sense.

The difference was stark. The first organisation treated legacy as a problem to be eradicated. The second treated it as a foundation to be modernised. One failed. The other thrived.

The Takeaway

The legacy trap is seductive because “replace everything” feels like progress. But in reality, it’s usually disruption in disguise.

Legacy is a constraint — but also an asset.
AI and automation can extend value while creating a bridge to the future.
Modernisation works best as evolution, not revolution.

Transformation leaders must remember: you don’t have to smash the legacy lock. You just need to pick it.

Breaking Change Fatigue

There’s a silent killer in transformation programmes. It’s not budget cuts. It’s not technology. It’s not even resistance to change.

It’s change fatigue.

That creeping exhaustion when teams feel like they’re on a never-ending treadmill of new initiatives, shifting priorities, and moving goalposts. It drains energy, erodes trust, and — if left unchecked — breaks transformations long before they deliver value.

Why Change Fatigue Happens

In almost every major programme I’ve been part of, change fatigue has surfaced at some point. The reasons are familiar:

Too much, too fast. Multiple change initiatives land on the same teams simultaneously. People are asked to adapt faster than they can absorb.
No stability between waves. Before one change embeds, the next one arrives. Wins aren’t consolidated, so employees feel they’re always running but never arriving.
Communication overload. Endless updates, but little clarity. People hear what’s changing but not why it matters or what stays the same.
Personal impact ignored. Leaders talk in terms of processes and systems, while employees feel the day-to-day disruption in workload, stress, and uncertainty.

Over time, the result is predictable: motivation dips, productivity falls, and talented people quietly disengage — or leave.

The AI and Data Dimension

AI introduces both risk and opportunity when it comes to change fatigue.

On the risk side, AI can feel threatening. “Will this replace me?” “Is my work valued?” If introduced poorly, AI becomes just another stressor.

But handled well, AI can reduce fatigue. By automating low-value tasks, it frees up people to focus on higher-value work. By analysing workload and employee sentiment, it can even help leaders spot burnout signals early and adapt change plans accordingly.

The key is positioning AI as a partner, not a disruptor. That requires trust — and trust requires transparency.

What Works Instead

Breaking the change fatigue spiral means treating change as a human experience, not just a business process. The organisations that succeed take a very different approach:

Sequence, don’t stack. Spread initiatives out. Create breathing space so employees can embed one wave of change before the next hits.
Celebrate wins. Don’t just move on to the next milestone. Take time to recognise progress and show people that effort is paying off.
Anchor communication in purpose. Move beyond technical updates. Share the “why,” connect it to strategy, and make clear what isn’t changing.
Position AI as an enhancer. Frame automation as removing drudgery, not removing people. Highlight how AI helps individuals succeed.
Listen and adjust. Use surveys, workshops, and analytics to capture how people are experiencing change — and adapt the roadmap in response.

A Story from the Field

A retail organisation I worked with launched three major initiatives in parallel: a new ERP system, an e-commerce platform, and a data modernisation programme. All three were important. But for frontline teams, it felt like chaos. New processes every month, conflicting training, and managers unable to answer basic questions.

Predictably, morale crashed. Staff turnover spiked. Productivity slumped. Leadership thought they had a technology problem. In reality, they had a fatigue problem.

In contrast, a healthcare provider I later supported took a phased approach. They sequenced initiatives deliberately: one wave of change, time to embed, visible celebration of success, then the next. They positioned AI tools not as replacements, but as helpers to reduce admin burden. Staff felt supported, not overwhelmed. Engagement scores rose. Adoption rates soared.

The difference wasn’t in the technology. It was in how people experienced the journey.

The Takeaway

Transformation isn’t just about designing new systems or processes. It’s about carrying people with you.

Too much change, too fast, breaks trust.
Wins must be consolidated before moving on.
AI can either add to fatigue — or reduce it, if introduced well.

Leaders need to remember: transformation isn’t a sprint. It’s a marathon. And marathons aren’t won by burning people out in the first few miles.

What Gets Measured Gets Done

There’s a saying in management that’s been repeated so often it risks sounding cliché: “What gets measured gets done.”

But in transformation programmes, it’s more than a saying — it’s a survival rule.

Because if you don’t measure the right things, transformation drifts. Effort gets wasted. Benefits evaporate. And leaders find themselves months (or years) in, with impressive slide decks but little to show in real business outcomes.

Why Measurement Goes Wrong

I’ve lost count of the number of transformations where measurement was an afterthought. The patterns are consistent:

KPIs defined too late. Teams launch projects, but only later scramble to figure out how to prove success.
Vague metrics. “Improved customer experience” sounds good in a business case — but what does it mean in hard numbers? NPS uplift? Churn reduction? Call handling time?
Focus on outputs, not outcomes. Organisations celebrate delivering a system on time and on budget — without asking whether it actually delivers business value.
Data gaps. KPIs exist on paper, but the organisation doesn’t have the data (or trust in that data) to track them properly.

The result is predictable: leaders declare victory at go-live, but the CFO quietly wonders why the P&L hasn’t shifted.

The AI and Data Dimension

AI has the potential to transform measurement — if used well.

Predictive analytics can forecast benefit realisation, highlighting early warning signals when projects drift. Real-time dashboards can show adoption, performance, and customer impact.

But AI also magnifies the problem if the foundations aren’t right. Feed it vague KPIs or poor-quality data, and you simply get faster, shinier reporting on the wrong things.

In other words: AI can supercharge measurement, but it cannot replace clarity of intent.

What Works Instead

Getting measurement right isn’t complicated — but it does require discipline from day one.

Define outcome-focused KPIs early. Ask: “What will success look like in business terms?” Not just “system deployed,” but “revenue uplifted,” “cost reduced,” “customer churn decreased.”
Connect KPIs to clean data sources. If you can’t measure it now, don’t assume you magically will later. Build the data pipelines as part of the transformation.
Track continuously, not just at milestones. Measurement should be real-time, not annual. Benefits realisation needs the same cadence as financial reporting.
Make ownership explicit. Every KPI should have an accountable leader who can influence it, not just a dashboard with no owner.
Use AI for early signals. Let predictive models highlight variance before it becomes failure. If adoption is lagging, if costs are trending up, leaders should know fast.

A Story from the Field

At a global insurer, I saw a digital transformation celebrated as a success: every system was delivered on time, every milestone ticked off. But months later, leadership admitted they couldn’t prove a single pound of value. Why? Because no one had defined what value actually meant — let alone set up a way to track it.

By contrast, at a telco I worked with, the transformation team flipped the approach. Before a single system was designed, they defined three hard KPIs tied directly to business value: reduce average customer onboarding time from 12 days to 3, cut churn by 5%, and lower cost-to-serve by 10%. Every project was mapped back to these.

The difference was night and day. Progress was clear, benefits were visible, and leadership had confidence that investment was paying off.

The Takeaway

Transformation without measurement is just activity.

KPIs must be defined early — and tied to business outcomes.
Data pipelines and trust in that data are as important as the systems being built.
AI makes measurement faster and smarter — but only if the foundations are clear.

Because in transformation, what gets measured gets done. And what doesn’t? It gets lost.

Continual Transformation

One of the biggest mistakes I see in organisations is treating transformation as if it were a project. A start date. A plan. A go-live. A party to celebrate delivery. And then… back to business as usual.

But transformation is not a project. It’s not something you “finish.” It’s a capability. A muscle. A way of working that has to be built, stretched, and sustained.

When organisations forget this, the result is predictable: momentum fades, old habits creep back, and within a year or two the shiny new model looks suspiciously like the one it was meant to replace.

Why Organisations Fall into the Trap

There are some familiar reasons transformation gets treated like a one-off project:

Funding models. Most organisations budget transformation as a capital investment with a fixed end date, not as an ongoing capability.
Leadership pressure. Boards and execs want milestones and certainty, not the ambiguity of “continuous change.”
Change fatigue. People long for stability, so leaders frame transformation as a phase to be endured, rather than a new reality to be embraced.
Consultancy playbooks. Too many external advisors still sell transformation as a 3-year roadmap, a neat programme that can be “completed.”

The problem is, markets don’t stop moving just because your project plan says “done.” Customer expectations evolve. Competitors launch new models. Technology shifts. Standing still is the fastest way to fall behind.

The AI and Data Dimension

This trap is even more dangerous in the age of AI.

AI models don’t just need to be deployed — they need to be trained, tuned, monitored, and adapted as data changes. What works today might degrade tomorrow if customer behaviour shifts or regulatory requirements change.

Treating AI adoption as a one-and-done project is a recipe for drift, bias, and loss of trust. Continuous oversight and improvement are non-negotiable.

The same applies to data. Data quality, integration, and governance are not “tasks to complete.” They are disciplines that have to be embedded into business-as-usual operations.

What Works Instead

The organisations that succeed treat transformation as an evolving journey. Their playbook looks more like this:

Build transformation into BAU. Don’t create a separate “change” team — embed change capability into line functions.
Fund continuous improvement. Move from project-based funding to ongoing investment models that recognise transformation is never done.
Measure and adapt. Use KPIs not just to prove success, but to spot where the next cycle of improvement needs to focus.
Treat AI as a lifecycle. Monitor, retrain, and evolve models continuously. Bake AI operations into the core, not as a side project.
Celebrate iteration. Shift the narrative from “we finished transformation” to “we are always transforming.”

A Story from the Field

At a financial services firm, leadership celebrated the delivery of a major transformation: new digital channels, new CRM, new data warehouse. The programme closed, the team disbanded, and BAU took over.

Within 18 months, cracks appeared. Data quality slipped. Customer adoption plateaued. The new CRM wasn’t being used as intended. When asked to fix it, the response was, “That programme ended.”

Compare that with a utilities provider I worked with. Instead of closing their transformation at “delivery,” they built a permanent Transformation Office. It wasn’t a programme team — it was part of operations. Their role was to keep strategy, delivery, and measurement aligned, continuously. The result? Every year, they layered on new capabilities, adapted processes, and improved data quality. Transformation wasn’t something they did. It was something they were.

The Takeaway

Transformation is not a one-and-done project. It’s a permanent capability.

Funding must reflect continuity, not just capital projects.
AI and data demand lifecycle management, not one-off deployment.
Leaders must build transformation muscle into the organisation’s DNA.

The most successful organisations I’ve seen don’t ask, “When will we finish transformation?” They ask, “How will we keep transforming?”

That shift in mindset makes all the difference.

Continual Transformation

But transformation is not a project. It’s not something you “finish.” It’s a capability. A muscle. A way of working that has to be built, stretched, and sustained.

When organisations forget this, the result is predictable: momentum fades, old habits creep back, and within a year or two the shiny new model looks suspiciously like the one it was meant to replace.

Why Organisations Fall into the Trap

There are some familiar reasons transformation gets treated like a one-off project:

Funding models. Most organisations budget transformation as a capital investment with a fixed end date, not as an ongoing capability.
Leadership pressure. Boards and execs want milestones and certainty, not the ambiguity of “continuous change.”
Change fatigue. People long for stability, so leaders frame transformation as a phase to be endured, rather than a new reality to be embraced.
Consultancy playbooks. Too many external advisors still sell transformation as a 3-year roadmap, a neat programme that can be “completed.”

The AI and Data Dimension

This trap is even more dangerous in the age of AI.

Treating AI adoption as a one-and-done project is a recipe for drift, bias, and loss of trust. Continuous oversight and improvement are non-negotiable.

The same applies to data. Data quality, integration, and governance are not “tasks to complete.” They are disciplines that have to be embedded into business-as-usual operations.

What Works Instead

The organisations that succeed treat transformation as an evolving journey. Their playbook looks more like this:

Build transformation into BAU. Don’t create a separate “change” team — embed change capability into line functions.
Fund continuous improvement. Move from project-based funding to ongoing investment models that recognise transformation is never done.
Measure and adapt. Use KPIs not just to prove success, but to spot where the next cycle of improvement needs to focus.
Treat AI as a lifecycle. Monitor, retrain, and evolve models continuously. Bake AI operations into the core, not as a side project.
Celebrate iteration. Shift the narrative from “we finished transformation” to “we are always transforming.”

A Story from the Field

Within 18 months, cracks appeared. Data quality slipped. Customer adoption plateaued. The new CRM wasn’t being used as intended. When asked to fix it, the response was, “That programme ended.”

The Takeaway

Transformation is not a one-and-done project. It’s a permanent capability.

Funding must reflect continuity, not just capital projects.
AI and data demand lifecycle management, not one-off deployment.
Leaders must build transformation muscle into the organisation’s DNA.

The most successful organisations I’ve seen don’t ask, “When will we finish transformation?” They ask, “How will we keep transforming?”

That shift in mindset makes all the difference.

Over-Promise, Under-Deliver

Every transformation begins with a promise. The promise of growth, efficiency, delighted customers, or new revenue streams. The bigger the promise, the easier it is to secure funding, board approval, and headlines.

But promises create expectations. And when delivery falls short — as it so often does — trust collapses. The organisation doesn’t just question the programme. It questions the leaders behind it. That’s the transformation cliff: the gap between ambition and reality.

Once you fall off, it’s hard to climb back.

Why Over-Promising Happens

Over-promising isn’t just about optimism. It’s baked into how organisations launch transformations:

Pressure to secure investment. Leaders feel they must sell big numbers to win approval. Modest returns don’t excite boards.
Vendor hype. Technology partners pitch best-case scenarios as guarantees. Leaders confuse marketing slides with delivery certainty.
Ignoring complexity. The sheer number of dependencies, integrations, and culture shifts are underestimated.
AI and automation hype. Too many assume AI will deliver magic results overnight, without data readiness or governance.

The effect is predictable: ambitious promises up front, painful under-delivery at the back end.

The AI and Data Dimension

AI magnifies this problem.

I’ve seen executives promise “AI will reduce costs by 30%” without checking if they even had clean, connected data. Models were built quickly but lacked accuracy. Adoption failed. Benefits didn’t materialise.

Data maturity is the silent deal-breaker. Without solid foundations — governance, quality, accessibility — AI can’t deliver at scale. Over-promising ignores this reality, and under-delivery is inevitable.

What Works Instead

Avoiding the cliff requires discipline and honesty from day one. Here’s what works:

Set expectations based on capability, not aspiration. Anchor promises in what the organisation can realistically deliver given current maturity. Ambition is fine, but reality first.
Assess data readiness before making AI commitments. No clean data = no credible AI benefits. Test this early.
Break down benefits into near-term and long-term. Don’t sell a single giant ROI figure. Show the curve: small early gains, bigger compounding value later.
Communicate uncertainty. Be transparent about risks, assumptions, and what might shift. Trust grows when leaders admit complexity, not hide it.
Track and re-baseline. Benefits aren’t static. Adjust projections when conditions change. It’s better to reset expectations than to deliver nothing.

A Story from the Field

At a retail bank, leadership promised that their “digital transformation” would cut operating costs by 25% within two years. Ambitious? Yes. Achievable? No.

They hadn’t factored in regulatory hurdles, legacy systems, or the skills gap in their workforce. Projects launched, but benefits lagged. At the two-year mark, costs had only dropped 5%. The board lost confidence. Investment dried up. The cliff edge had been reached.

Contrast that with a telco I worked with. Their leaders promised modest initial returns — a 3–5% efficiency gain in year one — but built a clear roadmap for scaling. As benefits landed, trust grew. By year three, they’d actually over-delivered, hitting double-digit cost savings. The difference? Realistic promises, underpinned by capability-based planning and honest communication.

The Takeaway

Over-promising might win funding, but under-delivering destroys trust. Sustainable transformation comes from realism, not bravado.

Anchor promises in current capability and maturity.
Test data readiness before committing AI outcomes.
Show benefits as a journey, not a jackpot.
Be transparent about risks and assumptions.
Re-baseline when reality shifts.

It’s not about lowering ambition. It’s about grounding ambition in reality. That’s how you build credibility, deliver value, and avoid the transformation cliff.

Why Lone Wolf Transformation Fails

Every organisation has its heroes.
The visionary leader who drives a transformation. The technologist who pushes through a new platform. The strategist who crafts the roadmap.

But here’s the brutal truth: lone wolf transformation doesn’t work.

When change depends on a single leader or a single team, it might start strong, but it rarely sustains. The energy fades, the organisation resists, and when the lone wolf moves on, so does the momentum.

Transformation, at scale, is a team sport. And the organisations that succeed build packs, not heroes.

Why Lone Wolf Approaches Fail

Over the years, I’ve seen this pattern play out:

Centralised control. Transformation sits in a single office or person. Everyone else is told, not engaged.
Lack of buy-in. Teams see the change as “the CIO’s project” or “the consultants’ thing,” not their future.
Speed over inclusion. Lone wolves believe that by working in isolation they’ll move faster. In reality, they create fragile designs no one wants to adopt.
AI and automation in silos. Data teams build clever models that never reach production because they’re not embedded into the business.

The result: when resistance hits, there’s no support base. When the leader leaves, the transformation dies.

The AI and Data Dimension

AI is particularly vulnerable to lone wolf thinking.

I’ve seen small innovation teams build fantastic proofs-of-concept — accurate models, automated processes, even predictive insights. But without business involvement, these stay in the lab. No adoption. No scaling. No value.

AI thrives on collaboration. It needs technical expertise, business context, data ownership, and change management. One team alone can’t cover it all.

What Works Instead

Successful transformations don’t rely on lone wolves. They build packs:

Cross-functional squads. Bring together business, tech, data, and operations. Transformation happens where these perspectives meet.
Shared ownership. Frame transformation as everyone’s future, not just IT’s. Success belongs to the business, not the project office.
Distributed leadership. Empower influencers across the organisation — not just at the top. People follow peers they trust.
AI adoption as collective effort. Engage data owners, business users, and process leads. Adoption is cultural as much as technical.
Transparency and co-creation. Share designs early. Invite contribution. Let people shape, not just receive, the change.

A Story from the Field

At a utilities company, a single exec drove a major transformation. He had the vision, the energy, and the authority. But he worked largely alone. Teams were told what was coming, not involved in shaping it.

When he left the business for another role, the transformation ground to a halt within months. With no shared ownership, there was no momentum to sustain it.

Contrast that with a healthcare organisation I worked with. They built transformation “guilds” — cross-functional groups of clinicians, IT, operations, and data specialists. The guilds co-designed new processes and platforms. When leaders rotated, the guilds stayed. Ownership was embedded across the business. The transformation didn’t just survive leadership changes — it thrived.

The Takeaway

Lone wolf transformation fails because it’s fragile. Shared transformation succeeds because it’s resilient.

Build packs of cross-functional teams.
Distribute ownership beyond IT or one leader.
Treat AI adoption as cultural, not just technical.
Co-create openly to build trust and adoption.

Because at the end of the day, transformation is too big, too complex, and too important for one wolf. But with a pack? You can go the distance.