Not when the results look good.

Not when the board agrees with them.

Not when the strategy they backed delivers exactly what they promised.

Watch what happens when the evidence starts going against them.

Some start rewriting the past until somehow, they were right from the beginning.

Suddenly it’s:

“I had concerns about this from the start.”

“We knew this could go wrong.”

And later, when it’s clearly failed:

“That was never meant to be the long-term plan anyway.”

Everyone remembers it differently. Nobody says so.

Others admit mistakes without ever saying what they actually did wrong.

“Mistakes were made.”

“We’ve learned from it.”

Then comes a promise that things will be handled differently next time.

It sounds like someone is taking responsibility. Except you still don’t know who made the call, what they got wrong, or who is going to do anything differently.

Usually, you never find out.

Then there are leaders who change their minds in public.

Clearly.

Without flinching.

They say:

“I backed this decision. The evidence now shows I was wrong. Here is what I missed, and here is what we are doing next.”

No rewritten history.

No vague collective apology.

No attempt to spread responsibility among people who did not make the call.

Being wrong is unavoidable.

Staying wrong is a leadership choice.

A CEO once told me about a senior hire he had personally fought to appoint.

Several members of his leadership team had reservations. He overruled them because the candidate had experience and an impressive record elsewhere. The CEO was convinced he had found the right person.

Six months later, the problems were obvious.

Targets were being missed.

Good people were leaving.

Customers were complaining.

Every leadership meeting brought another explanation for why performance would improve next quarter.

The CEO kept defending the appointment.

The market had changed. The role had proved harder than expected. The team had failed to support the new executive. The targets were unrealistic, etc.

There is some truth in each explanation. There is also a simpler explanation he kept avoiding.

He had made the wrong hire.

His leadership team already knew.

The board knew.

And, by then, the CEO knew too.

Removing the executive meant reversing a decision he had argued for with complete confidence.

So the business spent another six months waiting for a result few people still believed would arrive.

The cost went far beyond salary and missed revenue.

His team noticed what happened when one of his decisions went wrong.

They could see that his own decisions were harder to challenge. Once he had publicly backed something, changing it became harder too. People noticed, and became more careful about what they said.

One bad decision had started affecting how people behaved around him.

Most executives see consistency as a virtue.

People expect leaders to have conviction and hold their nerve rather than change direction whenever conditions become uncomfortable.

Fair enough.

But hold on to consistency for too long and it gets expensive.

The strategy you championed is not working.

Perhaps the acquisition you pushed through is destroying value. Or the new market is burning through cash and the customers simply are not there.

Maybe that senior hire is making the team worse.

Meanwhile, the numbers no longer support what you told everyone in January.

At that point, consistency no longer looks like strength.

It looks like attachment.

Once you are defending a position because it is yours, the job has changed. You are protecting your previous answer instead of looking for a better one.

Senior leaders have plenty of ways to disguise that, including from themselves.

They ask for more data even though the existing data already points somewhere uncomfortable.

Success criteria get moved.

A flawed strategy becomes an execution problem.

Critics get labelled negative or impatient.

Deadlines keep slipping because maybe the next quarter will prove them right.

And money already spent becomes another reason to spend more.

People notice.

They learn pretty quickly which subjects can be challenged and which opinions belong to the boss.

Evidence that might trigger a defensive reaction gets softened or left out. Recommendations become less direct. People wait for somebody else to say what everyone can already see.

Eventually, the CEO stops hearing the truth while still believing everyone feels free to speak.

Leadership is not about being right for the longest period.

It is about getting to the right answer soon enough to do something about it.

Saying “I got it wrong” is only part of that.

Good leaders tell people what changed their minds. That way, the team can see why the decision changed rather than wondering whether the leader simply lost their nerve.

Changing your mind because the evidence changed is very different from changing it because people in the meeting became uncomfortable.

The first is judgement.

The second makes people wonder what you will believe next week.

A change of position sounds something like this:

“We expected customer retention to improve once adoption reached 60%. Adoption is now above 70%, but retention has not improved materially. Our original assumption was wrong. We will stop investing in the current approach and test a different one.”

Nobody has to guess what happened. The assumption was clear, the result did not support it, and the decision changed.

That is enough. There is no need for the leader to make a show of admitting the mistake.

People around the CEO start behaving differently when they see this happen.

Product leaders find it easier to drop weak features they once argued for.

Sales leaders are less likely to keep defending a forecast that no longer makes sense.

Board conversations spend less time on who gets the blame and get to the next decision sooner.

A decision becomes something that can be tested, rather than something somebody has to defend forever.

In my experience coaching CEOs, a few habits make a noticeable difference.

→ Announce changes of mind clearly.

Say, “I have changed my mind on this. This is what changed it.”

People should not have to work out whether the strategy changed or whether they misunderstood what you said before.

Clarity protects execution.

→ Own the original call.

If you made the decision, say so.

Your team already knows. Dodging ownership simply gives them something else to distrust on top of the original mistake.

→ Separate the decision from the decider.

“That call was wrong” is useful.

“I am hopeless at this” is theatre.

Self-criticism can turn a discussion about a bad decision into everyone reassuring the person who made it.

Own the error and keep the meeting about the decision.

→ State the assumptions behind major decisions before results arrive.

For example:

“We expect this investment to produce £2 million in qualified pipeline within six months. If it produces less than £750,000, we will reconsider the channel.”

That gives everyone something concrete to come back to later. You decided in advance what would count as evidence that the bet was not working.

→ Revisit your three biggest decisions every quarter.

Try looking at each one as though it came from somebody else.

Go back to what you believed when you approved it and compare that with what you know now. Some assumptions may still hold. Others may have been proved wrong.

Then ask yourself whether you would approve the same decision today.

If you would not, why are you still living with it?

→ Reward the person who proved you wrong.

Do it publicly.

Say, “Sarah challenged my view early, and the evidence supported her position. I want more of this.”

You probably will not need to say it twice.

People remember what happens to the person who disagrees with the boss, especially when that person turns out to be right.

→ Change course before you have a perfect alternative.

Leaders sometimes keep funding a failing strategy because the replacement still has unanswered questions.

That is an impossible standard.

You can stop doing something that has failed without pretending you already know exactly what will replace it.

None of this means changing direction every time somebody disagrees with you.

Constant reversals exhaust teams and destroy confidence. A sound decision also needs enough time to work.

The question is whether the evidence still supports your position. If the evidence has gone and pride is doing the work instead, you have your answer.

Being wrong costs something.

Maybe money. Maybe time. Maybe some credibility.

But finding out you were wrong gives you information you did not have before.

Ignoring that information is a choice.

Your team does not expect every decision to be right.

They do expect you to be straight with them about what you knew at the time, what you assumed and what has changed since.

And they watch what happens to the people who challenge you.

Punish or exclude the challenger and people will remember. Asking for candour afterwards will not fix it.

Hear them out, especially when the evidence supports them, and people are more likely to tell you what you need to know next time.

The strongest leaders are not the ones who never change position.

Their reputation does not depend on proving yesterday’s answer was right.

When the evidence changes, they are willing to say so.

And then they act.

Most new or resetting CEOs waste their first 100 days trying to fix surface problems instead of diagnosing the underlying system. Run this 4-part operational audit to identify your highest-leveraged bottlenecks, rate your executive team without bias, and align your entire organisation around 3 unassailable priorities.

The 100-Day Trap

Taking over as CEO usually means inheriting a long list of things that already look urgent. Sales may be off target, a product launch may be late, and your calendar fills before you’ve had much time to understand what’s actually going on.

That makes it very easy to spend the first few months reacting. The better move is to use that time to work out which problems are symptoms and which ones are causing the damage. Fixing things before you know the difference can leave you with the same problems a few months later.

Here is the exact framework for auditing your organisation across four core pillars: People, Strategy, Execution, and Governance.

1. The Talent Audit: The 90-Day Standard

You inherit or build an executive team, but you rarely audit them objectively. Evaluate every C-suite leader and VP using The Keystone Talent Framework:

• High Performance + High Values Fit: The Growth Engine

• Action: Protect, reward and stretch them with expanded scope. They multiply the output of the people around them.

• Low Performance + High Values Fit: The Coaching Priority

• Action: Find out whether they’re in the wrong role or genuinely short on a skill. Give them a defined period to close the gap.

• High Performance + Low Values Fit: The Organisational Poison

• Action: Terminate immediately. High performers who damage the culture can drive out your best people faster than a competitor ever will.

• Low Performance + Low Values Fit: The Fast Exit

• Action: Exit within 30 days. Dragging it out rarely helps the person or the rest of the team.

2. The Strategic Audit: Signal vs. Friction

Bad ideas rarely break strategy; it breaks due to strategic bloat. Ask your top 10 executives to answer these three questions in writing independently:

1. What are the top 3 priorities of this business over the next 12 months?

2. What is our single biggest competitive bottleneck right now?

3. What is one project we should completely stop doing today?

If you receive 10 different variations of priorities, you do not have a strategy problem; you have an alignment failure.

A successful strategy audit reveals complete alignment across the executive layer. If your team cannot articulate the top 3 objectives in exactly the same way, your front-line employees are executing in total darkness.

3. The Execution Audit: Calculating Operating Velocity

Execution velocity isn’t measured by how busy your team looks—it’s measured by decision speed and meeting leverage.

[ Strategic Decision Made ] ──► [ Execution Started ] = Operating Velocity

Audit your operating system against these three operational friction points:

• Decision Latency: How many days elapse between identifying a critical issue and locking a decision? High-performing companies make non-reversible (Type 1) decisions in days and reversible (Type 2) decisions in minutes.

• Meeting Tax: Look at your calendar and your executives’ calendars. If leaders spend more than 60% of their week in status-update meetings, your communication architecture is broken.

• Metric Transparency: Can any employee in your organisation name the core company metric they directly influence? If not, accountability is non-existent.

4. The Governance & Cash Audit: Defensive Security

Before you commit to the plan, make sure the company can actually pay for it. You also need to know what your board thinks could go wrong. If their concerns are very different from yours, find out why before spending more money.

• Runway & Capital Efficiency: Re-evaluate your burn multiple (Net Burn ÷ Net New ARR). A burn multiple above 2.0 requires immediate cost restructuring regardless of current cash reserves.

• Board Agreement: Schedule 1-on-1 diagnostic calls with every board member. Ask: “What is the one risk you see that the management team is currently underestimating?”

• Compliance & Debt: Check whether you’re too reliant on one or two customers. Look for any legal issues that could cost the company money, and identify roles where losing one person would cause a serious problem.

The 100-Day Audit Playbook: Execution Timeline

• Days 1–30: LISTEN & DIAGNOSE

• Have 1-on-1 conversations with your top 20 leaders and a handful of individual contributors.

• Join the regular executive meetings. Spend more time listening than talking.

• Speak directly with your top 10 customers and main investors.

• Days 31–60: AUDIT & SYNTHESISE

• Go through every VP and C-suite leader using the Keystone Talent Framework.

• Send the three strategy questions to the executive team and compare their answers.

• Then get into the finances. Check the burn rate and look at how the company is actually spending its time week to week.

• Days 61–90: REALIGN & COMMUNICATE

• Present audit findings directly to the Board of Directors.

• Rework the C-suite where needed. Exit leaders who aren’t working out and promote the people who are ready for more.

• Publish the three priorities for the entire company.

• Days 91–100: LOCK IN THE OPERATING SYSTEM

• Roll out the new meeting rhythm, OKR framework, and monthly reporting cadence.

The objective of your first 100 days isn’t to make everyone feel comfortable—it is to bring absolute clarity to the organisation. Clarity creates speed; ambiguity creates friction. Audit early, decide quickly, and build the foundation that allows your company to scale predictably.

Compounding this, the leadership agenda was becoming crowded. Fundraising, brand architecture, category definition, international partnership opportunities and pipeline expansion were all competing for the same finite executive attention. Without a disciplined framework for prioritisation, the business risked dissipating its energy across too many fronts precisely at the moment when focus mattered most.

The Approach:

We applied a structured interrogation of the growth agenda, moving systematically through commercial traction, capital readiness, brand strategy and organisational design. Central to this was establishing a category-of-one positioning that the business could genuinely own, then aligning the brand architecture around it: a deliberate separation between the corporate, investor-facing identity and the customer-facing experience, tested against proven analogues from category-defining technology companies.

We then reframed the leadership conversation from fundraising as an end state to funding as a trigger event. Recognising that capital would expose rather than resolve structural fragility, we defined the post-investment priority as a rightsising programme across people, roles and focus. We also applied a disciplined filter to new market opportunities, distinguishing partnerships that compound the core strategy from expansion that would open unnecessary new fronts, and committed to sponsoring only those that passed the test.

The Outcomes:

• A defensible category-of-one market position, with brand architecture aligned to it across investor and customer audiences

• Double-digit improvement in core engagement metrics acknowledged and converted into fundraising narrative

• A clear post-funding operating agenda agreed in advance, turning organisational fragility from a hidden risk into a managed programme

• Sharpened strategic focus, with new partnership opportunities qualified against the core growth thesis rather than pursued opportunistically

If you are serious about growth - yours or your company’s - and would like to explore coaching, feel free to book an exploratory call.

Let's Talk

Compounding this, the business was cash-constrained at a critical inflection point. A pipeline of significant contracts was maturing on a three-to-six month horizon, but the gap between present resources and future revenue required bridge capital. The founder was seeking not just funding, but experienced strategic guidance from someone who had navigated the full arc of building, scaling, and exiting a high-growth technology business.

Our Approach:

Recognising that the founder’s instinct to niche down was strategically sound, we worked to sharpen the distinction between the core medtech proposition and the broader SaaS platform opportunity. Rather than treating these as competing priorities, we reframed them as sequenced value creation layers, each serving a different phase of the growth plan and a different investor thesis. This gave the founder a cleaner, more compelling narrative for capital conversations.

We also introduced a more rigorous lens on investor fit, moving the conversation away from passive capital towards identifying partners with genuine operational proximity to the challenges ahead. A warm introduction to a leading sector-specialist investor was facilitated, alongside a framework for evaluating advisory relationships based on stage relevance, contextual expertise, and chemistry rather than conventional credentials alone.

The Outcome:

• Strategic positioning clarified from a generalist technology provider to a focused medtech company, with a sequenced roadmap to a scalable SaaS product targeting a global addressable market.

• Investor pipeline activated through a targeted introduction to one of the UK’s leading SEIS-specialist investors, significantly shortening the path to a first close.

• Advisory relationship scoped on an independent, conflict-free basis, preserving the integrity of guidance and aligning incentives with long-term organisational success.

• Founder equipped with a clearer framework for evaluating future capital and advisory partners, grounded in expertise and operational relevance rather than experience alone.

If you are serious about growth - yours or your company’s - and would like to explore coaching, feel free to book an exploratory call.

Let's Talk

I see this all the time when I coach CEOs. The decision sits there for weeks. Everyone in the room knows what should happen. But instead of acting, someone says what sounds responsible and buys another month: “Let’s get a bit more data before we decide.”

So someone builds a deck. That spawns a request for a second opinion. Which leads to a three-option model, because one scenario looks thin. By the time that all comes back, the situation is different - and now the numbers are wrong. Fresh analysis is needed on a problem that has moved on. The decision itself never lands. It just keeps getting harder to avoid.

The pattern hiding inside diligence

Here’s what most leaders won’t admit. Smart people don’t procrastinate in obvious ways. They delegate.

A junior manager who wants to avoid a tough decision just ignores the email. A seasoned CEO in the same spot asks for a review. It looks and feels like careful work. It even gets approval. But really, both actions just delay having to commit to something that might be wrong.

The more senior you get, the more respectable your stalling looks. I had to learn that the hard way. For a long time I thought pulling in more information was the prudent thing to do. But there’s a line — and past it, gathering data isn’t making you smarter. It’s just buying you a few more weeks before you have to face the thing you already know.

It’s also worth noting that decisions that truly need more data rarely get stuck. Low-stakes, easy-to-reverse choices move quickly because there’s nothing scary about being wrong. It’s the big decisions, the ones you can’t undo, where suddenly every dataset must be perfect. Pay attention to how much analysis someone asks for. It usually shows how much they fear being the one who made a mistake.

What it’s actually costing you

The cost never appears as a line in the analysis. That’s why it’s so easy to overlook.

The candidate you were “doing due diligence on” took the other offer in week three of your deliberation. The acquisition you kept modelling got bought by someone faster, and now it’s a competitor. The underperformer everyone has been managing around for two quarters is still there, still being managed around, while your best operator quietly works out that nothing changes here and starts taking recruiter calls.

None of that shows up in the next spreadsheet. The spreadsheet only tracks what you’re studying. It never measures the cost of the study itself. So you focus on the visible numbers and pay for the hidden ones.

And honestly, the slower damage is worse. Your people are not oblivious. They sit in those rooms too. After watching the same decision come back for round three, they work out the rules of the game here. Don’t bring anything half-baked. Make it airtight or don’t bring it at all. So they stop raising things early — which is exactly when you’d have the most ability to do something useful. What starts landing on your desk instead is already over-prepared, over-defended, and often quietly wrong in ways nobody will say out loud.

The Information Stop Rule

Before you ask for another analysis, try these three checks. They take about thirty seconds and you can do them right in the meeting, out loud.

1. The change test. What would this new information need to show for you to make a different decision? Say it before you ask for the data. If you can’t name the result that would change your mind, you don’t need the data. You’ve already decided and are just looking for cover.

1. The waiting cost test. Forget general principles — price this specific delay out. What does one more week of not deciding cost you here? Be specific about it. The contract sitting unsigned. The hire that hasn’t started yet. The morale of whoever’s watching this go round for the third time. Write those down, then stack them against whatever extra certainty you’d genuinely gain. Usually the maths is uncomfortable. Act on it..

2. The honesty test. This is the one most people skip. Ask yourself this: if the data came back exactly the way I expect, would I actually do anything different? If the answer is no — if you already know what you’d do — then what you’re looking for isn’t information. It’s permission, or cover, or someone else’s name on the decision. That’s worth being honest about, because no amount of analysis is going to fix the real problem.

If a decision passes all three tests, go ahead and gather the data. You have a real gap and a clear stopping point. If it fails any of them, you’re not being thorough. You’re hiding. Make the call.

Think about the last big decision you sent back for “more work.” Run it through these three tests. I bet you’ll find it failed at least two of them when you first delayed it.

What’s one decision you’ve called “needing more data” that you could actually make this Friday?

They have solved the resourcing problem.

They have removed themselves from day-to-day execution.

This is, by any conventional measure, good management.

What they have not solved - and what nobody mentions - is the interpretation problem.

Your CFO's title says cash management. The harder function is making risk visible to someone who would prefer not to look at it.

Your COO's title says operations. The real job is translating your half-formed strategy into something a team can act on Monday morning.

Every C-level role is a translation layer between what you intend and what the company does.

That layer needs ongoing signal from you to function.

When you withdraw - because the experts tell us delegation is supposed to look like withdrawal - the layer doesn't pause.

It keeps moving.

On assumptions rather than instructions.

A COO who can't get 20 minutes of focused attention doesn't stop making decisions.

They start making guesses.

I've seen founders lose two C-suite leaders within three months.

Different reasons given. Same thing underneath.

They'd run out of information about what the CEO actually wanted.

So they left.

Building the team solved the capacity problem.

Stepping away from it created a different one.

Then it breaks.

Not dramatically. Quietly. The best performers start optimising for themselves. Incentives quietly diverge from outcomes. The team around them stops contributing because the contribution isn’t visible enough to matter. And the CEO is left managing a collection of individual performances rather than something that compounds.

The difference between a high-performing individual and a high-performing team isn’t talent density. It’s what the reward system actually measures.

Most companies say they reward team outcomes. In practice, they promote the person who closed the biggest deal, shipped the fastest, or got noticed in the room. Everyone else updates their mental model accordingly.

I’ve watched leadership teams with genuinely exceptional people produce genuinely average results. Because no one had ever built the connective tissue between them. No shared model of how decisions get made. No visibility on what everyone else is carrying. No reason to pull in the same direction when pulling alone was faster.

Great individuals in a bad system will eventually behave like the system tells them to.

You can’t hire your way out of that.

Marshall Goldsmith has a phrase that should unsettle every CEO. The superstition trap. It is a corrosive belief that, because a particular behaviour produced results in the past, it will continue to do so indefinitely. It sounds rational. It isn’t.

The logic runs something like this. I did X. I succeeded. Therefore, X caused my success. Therefore, I must keep doing X. It is the kind of reasoning that would embarrass a first-year philosophy student, yet it governs the strategic thinking of some of the most accomplished business leaders.

These leaders are, by any reasonable measure, extremely good at what they do. That is exactly what can make them so vulnerable. Success breeds repetition. Repetition breeds rigidity. And rigidity, in a market that reinvents itself every eighteen months, is a death sentence delivered on a delay.

Consider the technology sector. A founder who scaled a business in 2015 by hiring aggressively, spending freely on customer acquisition and chasing growth at all costs would, if they attempted the same playbook today, find themselves running out of runway before lunch. The conditions have changed. The capital markets have changed. The regulations have changed. The founder, too often, has not.

This is not a problem confined to Silicon Valley or Silicon Roundabout. It is endemic in most businesses worldwide. The instinct to repeat what worked before is, in evolutionary terms, perfectly sensible. Our ancestors survived by sticking with the watering hole that didn’t kill them. But a FTSE 250 is not a watering hole - however much certain management consultants might wish it were.

What Goldsmith is really describing IMO is a failure of self-awareness masquerading as confidence. The executive who insists on the same go-to-market strategy, leadership style, and Monday morning routine is not demonstrating conviction. They are unwilling to interrogate their own assumptions. There is a word for that. It is not leadership.

The antidote is rather unglamorous (and a bit boring). It is the willingness to sit with the discomfort of not knowing. To accept that what made you formidable at thirty-five may be making you irrelevant at fifty. To treat your own past performance not as a template but as a data set - useful, but incomplete, and increasingly out of date.

Business has never been short of talented people. What it has been short of, persistently, is talented people willing to abandon what they are good at in favour of what they need to become. That is the real cost of the superstition trap. Not that it produces failure. But it produces a very convincing imitation of competence, right up until the moment it doesn’t.

There is a particular kind of stall that happens in growing companies. Not the dramatic kind we all read about. Not a market collapse or a product failure or a funding crisis. A quieter and hard to see one. Revenue flattens. Hiring gets harder. The leadership team starts having the same conversations every quarter without anything materially changing. And everyone looks outward for the explanation. The market. The competition. The economy.

Almost nobody looks at the CEO.

Which is strange, when you think about it, because the CEO is the one variable that typically has stayed constant through every phase of the company’s growth. The product has changed. The team has changed. The strategy has probably been rewritten twice.

But the person making the final calls, setting the rhythm, deciding what gets attention and what gets ignored. That person has been there since the start, operating with roughly the same set of instincts.

This article is about a pattern I have seen repeatedly in my work as a CEO coach, across more than a hundred engagements with founders, CEOs, and leadership teams. The pattern is this:

The business grows until it hits the boundary of the CEO’s current capability, and then it stops.

Because the leader reached the edge of what they knew how to do (or were willing to do) and the organisation faithfully reflected that limit back to them.

I should be clear about something. This is not about bad leaders. The CEOs who hit this wall are typically smart, driven, deeply committed people who built something real from nothing. The problem is not about a lack of talent. The problem is that the role changed around them and they did not change with it. Or, more precisely, they did not change fast enough.

Harvard Business School research on founder-CEO succession suggests that replacement is common as companies scale, and many founders step aside before taking the company public. These are not failures. Many of these founders built valuable companies. They simply could not (or would not) adapt to what the company needed from them at the next stage.

The question I will try to answer is what that adaptation actually looks like. Not in theory. In practice. What specifically needs to change, why it is so hard to change it, and what happens to the companies led by CEOs who manage to make the shift versus those led by CEOs who do not.

Why Does Growth Stall When the CEO Hasn’t Changed?

Because the CEO not changing is the problem. The business evolves. New products, more customers, bigger team. But the person at the top is still operating with the same habits, the same instincts, the same tolerance for risk that got them to this point. And for a while, that works. It even feels like strength.

There is a well-documented pattern in scaling businesses where the founder or CEO becomes the single biggest constraint on growth. The Theory of Constraints, originally a manufacturing concept from Eliyahu Goldratt, applies with a rather uncomfortable precision here. The throughput of any system is limited by its tightest bottleneck.

In a company where every significant decision routes through one person, that person’s bandwidth, judgement, and willingness to let go determine how fast anything moves.

According to a McKinsey study on decision‑making, a large majority of senior executives report that decision‑making dynamics slow down organisational progress. Too many meetings, too little clarity about who actually had authority, too much waiting for sign-off from the top. Not because leaders were incompetent, but because the structure had never evolved past the early stage where centralised control made sense.

I see this constantly. A founder who was brilliant at getting to two million in revenue is now running a six-million-pound business with exactly the same operating rhythm. Same Monday meetings. Same approval chains. Same instinct to jump into client calls when something feels off. The business has grown but the leadership architecture hasn’t, and at some point the gap between the two becomes the thing that stops everything.

The tricky part, and this is where I think most advice on this topic gets it wrong, is that the CEO bottleneck rarely looks like a problem from inside.

It looks like high standards. It looks like caring. A case study of a founder‑CEO revealed that, in a single week, they were involved in dozens of decisions, most of which did not actually require their unique expertise at that stage.

Research suggests that founders often spend a large majority of their time on day‑to‑day operations, even after the company has grown beyond the early stage. Sit with that number for a moment. More than two-thirds of the CEO’s time going to work that someone else could do, while the work that only the CEO should do. The thinking, the relationship-building, the long-range positioning) gets squeezed into whatever is left over.

The business did not create that pattern. The CEO did. Usually without realising it.

What Limits Are CEOs Setting Without Realising It?

The limits that matter most are rarely strategic - they are behavioural. They show up in which decisions a CEO refuses to delegate, which meetings they insist on attending, and how much discomfort they can sit with when someone else does something differently than they would have done it.

Decision hoarding

The CEO remains the approval point for too many things. Pricing, hiring, client escalations, product changes. Sometimes deliberately, sometimes not. The result is the same: initiative dies slowly. People stop trying to make calls independently because they have learned that the CEO will weigh in anyway.

McKinsey’s research on decision‑making effectiveness suggests that large organisations can waste hundreds of thousands of managerial days annually due to poorly designed decision processes and unclear accountability. Scale that down to a 50-person company and the pattern is the same. Just more intimate and harder to hide from. The CEO who approves every hire, who signs off on every marketing campaign, who reviews every client proposal is not leading. They are creating a queue.

Identity lock

This one is harder to talk about because it sits closer to the person’s sense of self, their sense of value (and often meaning). The CEO who built the thing, who sold the first client, who wrote the original code. That person often cannot separate their identity from the operational work of the company. Letting go of the doing feels like letting go of who they are.

Research published in Administrative Science Quarterly found that founders invest their sense of self in how they understand and define their ventures. Entrepreneurial identity is tightly coupled with interpretations of responsibility, agency, and control.

This is not a personality flaw. It’s actually a rather predictable consequence of having built something from nothing. But it means that when the business needs the founder to step back from the work that made them feel like a founder, it triggers something deeper than a process change. It feels like loss.

I worked with a CEO last year who could articulate perfectly well that she needed to step back from product decisions. She had read the books. She had been to the conferences. And yet every time the product team made a call she disagreed with, she overruled them. Not aggressively. Quite politely, in fact. But the effect was the same.

Within three months, the product team had stopped making decisions without her. They would present options and wait. The form of delegation existed. The function of it did not.

Tolerance calibration

Every CEO has a threshold for how much mess they can tolerate. How wrong does someone need to get it before you step in? If your threshold is very low, if you correct course on small things, if you redraft the email your marketing lead wrote, if you jump onto a support call because the tone was slightly off, then you are teaching the entire organisation that your version of perfection is the only one that counts.

The irony is that this tends to get worse as the company gets bigger, not better.

More people means more decisions being made that the CEO cannot see, which creates anxiety, which creates more intervention, which creates more dependency. A Bain study on scaling found that firms which sequence improvements correctly achieve 30% higher ROI compared to those that skip straight to adding resources. The discipline is not about working harder. It is about working on the right constraint, and in most cases the constraint is the CEO’s own behaviour.

None of these are character flaws, by the way. They are design flaws. The habits that worked when the company was five people around a table become the exact mechanisms that prevent it from operating at fifty or five hundred.

How Does Your Comfort Zone Become the Company’s Ceiling?

Through repetition, mostly. The CEO’s habits set the operating tempo for the entire organisation, and over time the organisation moulds itself around those habits until they become structural.

Think about how information flows in a company where the CEO wants to be across everything. People learn to brief upwards before acting. Meetings become status updates rather than decision-making sessions. The leadership team develops a kind of learned helplessness. Perfectly capable individuals who have been conditioned to check before committing.

Research in team psychology suggests that shared mental models, trust, and collective orientation strongly influence how well teams coordinate and solve problems. When a CEO’s behaviour signals that the team’s judgement is not fully trusted, those cognitive states erode. People become more cautious, less creative, slower. Not because they are less competent, but because the environment has taught them to wait for permission.

There is a compounding effect here as well that is easy to miss. The CEO who stays too involved creates a team that requires more involvement, which reinforces the CEO’s belief that they need to be involved. It is a loop, and once it is established, it runs on its own. I have seen this play out dozens of times. The CEO says “my team cannot make decisions without me” and they are right. But they are also the reason why.

The Scale Up Institute’s 2023 Annual Review highlights that leadership development and capacity‑building are widely seen as critical for sustained growth, and that companies investing in their leadership teams’ decision‑making capability tend to be more likely to maintain high‑growth trajectories.

The constraint was not market access or funding. It was leadership bandwidth. And in most cases, the CEO was the one setting the bandwidth limit.

The information problem

A long‑running leadership study by Merryck & Co. and the Barrett Values Centre found that leaders’ self‑assessments often differ significantly from how their peers see their strengths and weaknesses.

This matters enormously for the CEO bottleneck problem because it means the CEO is the last person to see it. Everyone around them can see it. The leadership team, the board, sometimes even mid-level managers. But the CEO can sometimes to using a completely different map. Thinking the constraint is the market or the team or the timing. The team knows the constraint is the person at the top but cannot say so, because the power dynamic makes that conversation almost impossible to have without an external catalyst.

I had a conversation recently with a CEO who told me, in slightly different words, that every time he stepped away for a week the business slowed down. He meant it as evidence that he was essential. I heard it as evidence that the organisation could not function without him. Which is not the same thing at all, though it took a few more conversations before he saw it that way too.

Why Do CEOs Resist the Shift Even When They See the Problem?

Because seeing the problem and feeling ready to fix it are completely different things. The shift from operator to leader requires dismantling habits that feel like strengths, and most people will not voluntarily give up what got them here.

Let me be specific about what the resistance actually looks like in practice, because it is rarely super clear or even dramatic. It is not the CEO who says “I refuse to change.” It is the CEO who agrees with every principle of delegation and empowerment and then, in the next meeting, overrides a decision their VP just made. It is the CEO who hires a COO to take operational load off their plate and then spends six months second-guessing them. It is the CEO who reads a book about letting go and then checks Slack at 11pm to make sure the client email went out correctly.

The gap between knowing and doing is where most CEO development stalls. And the reasons are not mysterious< They are unfortunately very predictable.

First, there is the competence trap. The CEO got to where they are by being good at operational work. Strategy, selling, product development, problem-solving under pressure. These are real skills, and they have been rewarded for decades. Asking someone to stop doing the thing they are best at is asking them to feel less competent, at least temporarily. That is an emotionally difficult request even when the intellectual case is clear.

Second, there is the speed illusion. When a CEO steps in and makes a decision, things move faster. Immediately. Visibly. The team sees the problem resolved and the CEO feels effective. But this is a local optimisation that creates a global slowdown. The decision was faster, yes, but the ten other decisions that were queued behind it, waiting for the CEO’s attention, are still waiting. And the team just learned that the fastest way to get something resolved is to escalate it.

Research on CEO behaviour suggests that stronger leadership practices are associated with higher sales growth, though the exact magnitude depends on context.The direction matters. CEOs who moved towards enabling their teams outperformed those who stayed in operational mode.

Third, and this one is often a very hard one to acknowledge. The fear of irrelevance. If the business can run without me, what is my job? For a founder especially, this question is a really tough one. The identity of the person is so intertwined with the day-to-day of the business that removing themselves from the operations feels like removing themselves from the story. It is not logical. It is emotional. And emotions, in my experience, beat logic in the short term almost every time.

I have lost count of the number of CEOs who have told me some version of “I know I need to let go” in one breath and then described exactly why they cannot in the next. The knowing is not the hard part. The doing is.

What Actually Changes When a CEO Confronts Their Own Limits?

The organisation speeds up - but not immediately, and not without a period that feels worse before it gets better. This is where most CEOs quit the process, and it is worth being direct about why.

When a CEO starts to pull back from operational decisions, there is typically a dip. The team makes mistakes. Decisions take longer because people are not yet confident making them independently. The quality of output drops slightly. This is real, and it frightens most leaders into reverting to old patterns. Research on leadership transitions suggests that companies often experience a short‑term dip in efficiency as teams adjust to new decision‑making patterns, but those that persist through this period can achieve significantly higher growth rates over the following year.

I think of this as the valley. You have to cross it. There is no way around it. And most CEOs, when they feel the dip, assume the delegation is not working rather than recognising that the dip is evidence that it is working. The team is learning to operate without the safety net.

What the shift looks like in practice

It looks like specific, mundane changes. Moving from daily check-ins to weekly ones. Setting a decision threshold. Anything under a certain value or risk level, the team handles without escalation. Letting the marketing lead publish the blog post without the CEO reading it first. Allowing a hire to go through with only the hiring manager’s approval. Small things that feel enormous to the person giving them up.

In my experience, the CEOs who make this transition well tend to do three things.

They get clear on what only they can do. This typically comes down to four areas: strategy, capital allocation, key external relationships, and culture. Everything else, and I mean everything, should be moving towards the team. This is harder than it sounds because most CEOs have opinions about everything and confuse having an opinion with needing to make the decision.

They build a decision-making framework. Not a complicated governance structure. Something simple. McKinsey’s work on decision categorisation is useful here. They distinguish between big-bet decisions that need CEO involvement, cross-cutting decisions that need a process, and delegated decisions that should be handled by individuals or small teams. Most companies have never made these distinctions explicit, which means everything defaults to the CEO.

And they find someone who will tell them the truth about whether they are actually letting go or just performing the appearance of it. A coach, a board member, a peer, someone with permission to say “you approved that meeting schedule but you are still attending all eight of those meetings and it is undermining the entire point.”

That last point matters more than it sounds. A lot of CEOs go through the motions of delegation without actually releasing control. They delegate the task but retain the veto. They say “you decide” and then visibly disagree with the decision. The team picks up on this within days and reverts to seeking approval. The form changes but the function stays the same.

How Do You Know If You Are the Constraint?

If the business slows down when you step away, you already have your answer. But most CEOs do not test this because they are afraid of what they will find. Or, more commonly, they never step away long enough for the pattern to become visible.

The diagnostic questions

There are a few questions I use with CEOs I coach that tend to surface the truth fairly quickly.

How many decisions in the last week required your specific expertise versus your general approval? Most CEOs, when they actually audit this, are startled by the ratio. The approval decisions. The ones where someone just needed a yes or no, can massively outnumber the ones where the CEO’s particular knowledge was needed.

When was the last time your team made a significant call without consulting you, and how did you react? The honest answer here is often revealing. If the CEO felt anxious, left out, or immediately reviewed the decision to check whether it was right, that tells you something about the operating dynamic. Not the team’s competence, the CEO’s comfort.

If you disappeared for three weeks, what would break? And, this is the critical follow-up, is that a source of pride or concern? I have had CEOs answer this question with something close to satisfaction. “The whole place would fall apart.” They mean it as evidence of their value. It is evidence of a design flaw.

What do your direct reports complain about when you are not in the room? If you do not know the answer to this, that is itself an answer.

The calendar test

There is a simpler diagnostic that does not require any self-reflection at all. Look at your calendar for the last two weeks. Categorise every meeting and every block of focused work into one of three buckets: work that only the CEO can do, work that the CEO is doing because no one else has been set up to do it, and work that someone else could do right now if the CEO would let them.

In my experience, the first bucket is typically 15 to 20 percent of the CEO’s time. The second bucket. The “no one else has been set up” category, is where the real opportunity sits. That is not a delegation problem. That is a system design problem. The CEO has not built the infrastructure for anyone else to take ownership.

The exam question for any CEO reading this is not “am I a good leader?” That is too vague and too kind. The better question is: has your organisation outgrown your current way of leading it? And if the answer is yes, or even maybe, the next question is whether you are willing to change before the business forces you to.

What Separates CEOs Who Break Through From Those Who Don’t?

The willingness to be temporarily worse at leading in order to become permanently better at it. That is a harder trade than it sounds, and most CEOs underestimate what it asks of them.

Over the years I have watched a clear pattern emerge between CEOs who successfully make this transition and those who get stuck. The difference is not intelligence, or even self-awareness, though both help. It is a specific kind of tolerance for discomfort combined with a willingness to have their assumptions challenged by someone they trust.

The CEOs who break through tend to share a few characteristics, though not all of them, and not in any tidy order.

They treat their own development as seriously as they treat business strategy. Not as a nice-to-have. Not as something they will get to when things calm down. As an operating priority. The Scale Up Institute data is clear on this: companies that invest in leadership development at the top outperform those that do not, by a margin that is hard to ignore. But most CEOs interpret “leadership development” as something they should provide for their team, not something they need for themselves.

They have built at least one relationship where the truth can be spoken without consequence. A coach, a mentor, a board member, a peer group. Someone who is not on the payroll and does not need to manage the CEO’s feelings. This matters because the feedback the CEO needs is almost never available inside the company. The power dynamic makes it too risky for direct reports to say “you are the bottleneck” even when everyone knows it. External accountability is not a luxury. It is a structural requirement.

They are willing to watch their team make decisions they disagree with and let the outcomes play out. This is perhaps the hardest one. Not intervening when you believe you know the better answer requires a kind of discipline that runs against every instinct the CEO developed in the company’s early stages. But it is the only way the team learns to carry the weight. And sometimes , not always, but sometimes, the team’s decision turns out to be better than what the CEO would have chosen. That is a good day, even though it does not always feel like one.

They accepted the dip. They crossed the valley. When efficiency dropped temporarily, when the team stumbled, when a client issue took longer to resolve because it did not go through the CEO’s hands, they did not panic and revert. They held the line, adjusted where needed, and let the new system find its rhythm.

The CEOs who do not break through

The pattern on the other side is equally clear, though harder to describe without sounding harsh. CEOs who stay stuck typically do one of three things.

They agree with the diagnosis and change nothing. They read the books, attend the workshops, hire the coach, and continue operating exactly as they always have. The language changes but the behaviour does not. Their direct reports learn to nod along during the “empowerment” conversations and then continue bringing every decision to the CEO, because the CEO’s actions have not matched their words.

They delegate and then sabotage. They hand ownership to the team and then erode it through small, continuous interventions. A question here, a suggestion there, a “just checking in” that everyone understands is actually a review. The team gets the message. Ownership is nominal. Authority is real only when the CEO agrees with the outcome.

Or they blame the team. The delegation did not work because the team is not good enough. They need better people. They need more experienced leaders. Sometimes this is true, sometimes the team genuinely lacks the capability for the next stage. But more often than not, the team has not been given a real chance to demonstrate what they can do because the system was never truly redesigned. The CEO shuffled the org chart but did not change the operating model.

The business reflects all of this faithfully. Quarter after quarter. The same revenue plateau. The same operational drag. The same conversations in the same meetings. The same feeling of working harder for diminishing returns.

Finally

The limit is the point at which the CEO decides they have grown enough, whether they say it out loud or simply stop doing the work of growing. The business will faithfully reflect that decision back to them.

I do not say this to be unkind. I say it because, after thirty years of watching businesses grow and stall and grow again, the pattern is too consistent to ignore. The companies that break through growth ceilings are almost always led by someone who was willing to be uncomfortable. Who looked at the way they had always led and said “that is not going to be enough for what comes next” and then did the genuinely hard work of changing.

The CEOs who make that choice do not always get it right first time. The transition is messy. There are false starts, reversions, moments where old habits reassert themselves under pressure. But the direction of travel matters more than the perfection of the execution.

And the CEOs who do not make that choice? They are not bad leaders. They built real things. But they will watch, eventually, as the business either stalls at their limit or brings in someone else to lead it past the point where they could not.

The choice is yours. But it is a choice. And pretending it is not, pretending the constraint is somewhere other than where it actually sits, is the most expensive form of avoidance I know.

The most dangerous CEO behaviour rarely looks like control.

It looks like support.

A founder steps in to save a difficult customer.

They rewrite the board pack at midnight.

They rescue the sales call.

They smooth over tension between two senior leaders.

They make the hard decision because everyone else is circling it.

And for a while, it works.

Revenue moves.

Investors relax.

The team feels protected.

The CEO feels useful.

Then the hidden cost arrives.

→ The team stops building judgement
→ Senior leaders learn escalation faster than ownership
→ Problems travel upwards instead of across
→ The CEO becomes the quality control layer
→ Speed depends on one nervous system

This is the part no one admits publicly.

Many CEOs are not overworked because the business needs them everywhere.

They are overworked because the business has been trained to wait for them.

I have seen this pattern inside small founder led companies and large complex organisations.

The scale changes.

The behaviour does not.

The CEO thinks they are protecting the business from mistakes.

They are often protecting the team from the reps that would build capability.

The shift is uncomfortable.

Because the next stage of leadership is not doing harder work.

It is letting other people carry harder consequences.

That means:

→ Letting the commercial lead own the messy customer conversation
→ Letting the COO defend the operating plan without rescue
→ Letting the leadership team sit in unresolved tension long enough to solve it
→ Letting a decision be 80 percent yours, not 100 percent the CEO’s

This is not abdication.

It is capacity building.

A business does not scale because the CEO works harder.

It scales when the people around the CEO stop needing permission to think.

The uncomfortable truth:

If every important decision still needs the CEO’s fingerprints, the company has not built a leadership team.

It has built a queue.

And the CEO is standing at the front, wondering why everyone keeps waiting.

Every leader who survived the last wave of automation did the same thing. They redefined what leadership meant before someone else did it for them.

Not after the board intervened. Not after the restructuring memo. Before.

That pattern is repeating right now. And most CEOs are on the wrong side of it.

The instinct when new technology arrives is to manage it. Contain it. Delegate it to a transformation team and review the dashboard quarterly.

That worked when automation was confined to the factory floor or the back office.

It does not work when the technology touches strategy, decision-making, and judgment.

This wave is different.

It is not replacing tasks. It is compressing the time between signal and decision.

It is reshaping what organizations need from the person at the top.

And the leaders who wait for clarity before acting will find that clarity arrives in the form of a board conversation they didn't initiate.

Here's what I've seen consistently in my work with CEOs running complex, large-scale organizations.

The ones who thrive through disruption don't start with the technology. They start with a harder question.

What does leadership need to mean now that it didn't need to mean eighteen months ago?

Not in theory. In practice. In how they spend their time, what they delegate, what they hold close, and what they let go of entirely.

Most leaders skip this step. They adopt the tools, restructure the teams, and keep leading the same way.

That gap between new capability and old leadership behavior is where stagnation takes root.

The technology moves. The operating model moves. But the leader stays fixed.

And fixed leadership inside a shifting system creates drag that compounds quietly until it becomes visible to everyone except the person at the center.

The leaders I've coached through previous waves, whether it was digital transformation, platform shifts, or supply chain automation, shared one trait.

They treated the disruption as a prompt to re-examine their own role first.

Not the org chart. Not the tech stack. Their role.

They asked what decisions only they should make now. What decisions the system could make better.

Where their attention created the most value versus where it was simply habit.

That willingness to redefine yourself before the situation forces it is not common IMO.

But it is the difference between leading through a transition and being reorganized by one.

The technology is never the threat. The threat is assuming that what made you effective before will keep you effective now.

P.S. I work with CEOs to pressure-test their leadership model against the disruption already inside their organization. DM me if you want to talk through what that looks like before your board does.

You cannot scale your company until you scale yourself.

The Short Version

• Culture is not something that happens to a company. It is a series of choices the CEO makes, or fails to make, every week.

• The behaviours you tolerate in your senior team become the operating standard for everyone beneath them. It does not matter what your values statement says.

• The cumulative cost of this tolerance. In attrition, in lost velocity, in the slow erosion of your best people’s willingness to care. It’s what I call the Tolerance Tax. Every company is paying it. Most have never calculated the bill.

• Fixing culture is not a programme. It starts with one question, asked every Friday: what did I let slide this week, and what did that signal?

Every CEO I have coached believes they have a culture

Most of them are wrong. What they have is a values statement and a set of behaviours they have learned not to look at too closely.

I sit in on board meetings, leadership offsites, quarterly reviews. The stated values are almost always fine. Growth mindset, customer obsession, transparency. That sort of thing. Printed on lanyards, painted on walls. And then you watch the Monday morning leadership meeting actually run, and you see something different.

The VP of Engineering who talks over the product lead every single time. Nobody says anything.

The sales director who consistently overpromises to clients and then drops the problem on the delivery team. Everyone knows.

The CFO who sends emails at 11pm and then wonders why the finance team looks exhausted.

These are not culture problems in the abstract. These are specific behaviours, repeated in specific rooms, with the CEO present or aware. And the fact that they continue is itself a decision.

Here is the uncomfortable part. The cumulative effect of repeated tolerance is indistinguishable from endorsement. Your team cannot tell the difference. And after a while, neither can you.

Most CEOs I work with do not think of tolerance as a choice. They think of it as patience, or pragmatism, or picking battles. Sometimes it is. But there is a line where patience becomes permission, and most leaders cross it without noticing because nobody tells you when you have crossed it. That is the nature of the thing.

The Tolerance Tax

I use this term because it reframes the problem in language CEOs actually respond to. Culture drift is not a soft issue. It has a number attached to it. Most companies just have not done the maths.

I worked with a HealthTech business a couple of years ago where the CEO was genuinely baffled by high attrition in engineering. Exit interviews pointed to culture. His response was to hire a Head of People and task them with fixing it.

Reasonable instinct. Wrong diagnosis.

The actual problem was that the VP of Engineering ran his function like a personal territory. He took credit for his team’s work in every board meeting. He punished dissent by reassigning people to dead-end projects. The CEO knew all of this - broadly - but kept describing it as a personality thing. A style difference.

What he was actually doing was choosing that culture. Every week he did not address it, he was reinforcing it. And no Head of People can fix a problem that the CEO is actively sustaining.

When they finally measured the cost, it looked like this. Seven senior engineers left in 14 months. At a conservative estimate of 1.5 times annual salary per replacement, that is somewhere north of £700,000 in direct recruitment and onboarding costs alone. Lost product velocity across three sprint teams for the better part of a year. Institutional knowledge that walked out the door and never came back.

But the real Tolerance Tax was harder to quantify and more damaging. The remaining team had learned the rules. Keep your head down. Do not challenge upwards. Protect your own work. That became the operating culture - not because anyone designed it, but because the CEO allowed it.

Here is the exam question: if someone showed you an invoice for £700,000 labelled “cost of not having one difficult conversation,” would you pay it?

You already are. You just have not opened the envelope.

Why the CEO is always the last to know

You will typically hear about culture drift after the damage is done. That is how hierarchy works. Pretending otherwise - the open-door policy, the anonymous survey, the all-hands Q&A - sounds good on a slide but rarely holds up in practice.

There are earlier indicators, though. Across roughly 100 coaching engagements, I have found four signals that show up reliably, in roughly this order.

Signal 1: People stop disagreeing with you in meetings.

Not because they agree. Because disagreeing carries an unspoken cost. You will not notice this happening because the meetings will actually feel more efficient. Shorter, fewer arguments, quicker decisions. It feels like progress. It is usually the opposite.

I missed this one myself, years ago, sitting on a board where the consensus felt comfortable. It took a non-executive pulling me aside after a session to point out that three people in that room had stopped contributing anything beyond confirmation. I had read silence as alignment. It was resignation.

Signal 2: Your best performers start leaving, and the reasons they give are vague.

Family reasons. A better opportunity. Wanted a change. The real reason - which they will tell a trusted colleague but never you - is usually something closer to: I stopped believing anything would change.

Signal 3: Strong new hires flame out fast.

This one is easy to misread. The instinct is to blame the hire, or the recruitment process. But if you are consistently losing good people within 12 months, the environment is rejecting them. They do not fit the actual culture, which is different from the culture you think you have.

Signal 4: Middle management starts managing around problems rather than escalating them.

This is a late-stage signal, and by the time you see it, the distance between your stated culture and your real culture has become structural. Your middle layer has learned to absorb dysfunction rather than surface it. They are not being disloyal. They are being rational. They have watched what happens when someone raises a problem, and they have drawn the obvious conclusion.

Whose job is this?

Both the CEO’s and the organisation’s. But the order matters, and this is where I see leadership teams get tangled up.

You can distribute the maintenance of culture. Team leads who reinforce norms, People teams who flag risks, rituals that keep things visible. All of that is useful.

But the initial standard - the line that says this is acceptable, this is not - has to come from the CEO. Specifically, it has to come from what the CEO does when the line gets crossed. Not what they say about it in an all-hands. Not what is written in the handbook. What they actually do in the room when it happens.

Think of it this way. The CEO sets the pressure at the mains. If the pressure is strong and consistent, every floor in the building gets water. If the pressure is weak or intermittent, the top floors might be fine - they are closest to the source - but by the time you get to the ground floor, nothing comes out of the tap.

The people furthest from the CEO feel the culture gap first. They always do.

So yes, distribute the maintenance. But own the standard. And own the consequences when it slips. That second part is where most CEOs quietly step back. And it is exactly where the damage begins.

The Three Moves

Most advice on culture is either too abstract to act on or too programmatic to stick. What follows is not a programme. It is three specific things you can do, starting this week, that will materially change how your culture operates. I have seen each of these work across dozens of businesses. The order matters.

Move 1: The Friday Reckoning

Ten minutes. End of the week. Before you close your laptop.

Three questions:

1. What did I tolerate this week that I should not have?

Write down the specific behaviour, the specific person, and the specific meeting or moment. Vague does not count. “The leadership team was a bit off” is not an answer. “James talked over Sarah twice in the product review and I said nothing” is.

2. What did that signal to the team?

Not what you intended. What they saw. If the VP of Engineering dismissed a junior developer’s concern and you moved on to the next agenda item, the signal is: hierarchy matters more than contribution here. If the sales director missed a client commitment and you discussed it privately rather than in the team meeting, the signal is: commercial performers play by different rules.

3. What am I going to do about it on Monday?

Name the conversation. Put it in your calendar. If you will not do it Monday, name the date you will. If you cannot name a date, you have just answered question one for next Friday.

This works because it is private, it is regular, and it creates a feedback loop you cannot ignore. No programme required. No offsite. Just a weekly, honest audit of the gap between what you say matters and what your behaviour actually reinforces.

Move 2: The Specific Conversation

This is the one most CEOs would rather avoid. The direct, private conversation with the specific person about the specific behaviour.

Not a team-wide email about values. Not a subtle hint in a one-to-one. Not a restructuring that moves the problem into a different reporting line. A conversation that names the behaviour, explains the impact, and makes clear what needs to change.

The structure is simple: I have noticed [specific behaviour]. The impact on the team is [specific consequence]. What needs to change is [specific expectation]. I am telling you this because I want you to succeed here, and this pattern is getting in the way of that.

Most CEOs I work with would rather restructure a division than have this conversation. I understand why. But the restructuring does not fix the problem if the person is still there doing the same thing in the new structure.

One thing I have learned from sitting on both sides of this. The conversation is almost never as bad as the anticipation of it. And the cost of not having it - which you have been paying, quietly, for months - is almost always worse than you think.

Move 3: Confront the Paradox Performer

This is where it gets genuinely difficult, and where I will not pretend there is an easy answer.

Some of your best commercial performers may also be your worst cultural actors. The sales director who closes the biggest deals but treats the operations team like a personal support function. The co-founder who built the product but cannot let anyone else make a decision about it. The technical lead whose brilliance is matched only by their ability to make everyone around them feel inadequate.

These are the cases where tolerance feels rational, because the short-term cost of confrontation is visible and immediate - a quarter of missed revenue, a product delay, a co-founder dispute - and the long-term cost of inaction is not.

But the long-term cost is there. It is always there. You just have not measured it yet.

Here is what I ask CEOs to do. Run the numbers. Take your paradox performer and calculate what their behaviour is costing you in the people around them. Attrition in their team. Recruitment costs. The productivity drag on the people who have to work with them. The ideas that are not being raised because people have learned it is not worth the fight. The senior hire who turned down your offer because they met this person during the interview process and drew their own conclusions.

When you put a number on it - even a rough one - tolerance stops being a judgement call and becomes a financial decision. And most CEOs are better at financial decisions than they are at cultural ones. Use that.

Can you rebuild culture once it has drifted too far?

Yes. It takes longer than you want, and it requires you to go first.

Going first means admitting - at least internally, and often to your team - that the current culture is one you helped build through inaction. That is not a comfortable thing to say. But it is the price of entry.

The questions I get asked most

How do I tell the difference between tolerating a behaviour and picking my battles?

Picking your battles means you have consciously decided that a specific issue is lower priority right now, and you have a rough timeline for when you will come back to it. Tolerating means you have noticed a behaviour, felt uncomfortable, and decided not to act - without setting a return date.

The test is simple. If you cannot name when you plan to address it, you are tolerating it.

What if the person I need to confront is a co-founder, or someone the board considers essential?

The logic holds regardless of seniority. If the behaviour is corroding the team, not addressing it is still a choice with consequences. In my experience, boards are more receptive to these conversations than CEOs expect - particularly when you bring specific examples and a clear framing of the business impact. Vague complaints about culture will not land. Specific patterns with measurable costs will.

Is a culture offsite ever useful?

It can be, but only if you have already done the hard work of identifying what needs to change and you are willing to hold people to it afterwards. An offsite that generates a list of values is fine as a starting point. The problem is that most companies treat the offsite as the finish line. The real work is the 50 weeks that follow it, and specifically what happens the first time someone violates the new standard.

How quickly can culture actually change?

Faster than most people think, but slower than most CEOs want. Visible shifts start appearing within three to six months if you are consistent. Full cultural change - self-sustaining, holds without you policing it - takes 12 to 24 months.

What if I am the one whose behaviour needs to change?

Then you are ahead of most CEOs, because at least you are asking the question. The answer is the same. Name the specific behaviour, understand what it signals, and change it visibly enough that your team notices. A quiet private adjustment does not work here. Your team has been watching you operate a certain way for months, maybe years. They need to see the change to believe it.

That does not mean a dramatic announcement. It means doing the thing differently, consistently, in front of the people who need to see it.

The Tolerance Audit

Before you move on, try this. It takes five minutes and will tell you more about your culture than any survey.

List your five most senior leaders. For each one, answer four questions:

1. What behaviour am I currently tolerating from this person?

2. How long has it been going on?

3. What is the business cost - in attrition, in team performance, in opportunities not pursued?

4. What conversation do I need to have, and when will I have it?

If you cannot fill this in for at least two of your five, you are either running an unusually healthy leadership team or you are not looking closely enough. In my experience, it is rarely the first one.

The exam question for any CEO reading this is not whether your culture is perfect. No culture is. It is whether you know, specifically, where it is drifting. Whether you can name the behaviour, name the person, and name the cost.

And whether you are willing to do something about it before the exit interviews tell you what everyone else already knew.

In 2026, velocity is a commodity. With AI agents managing 70% of operational execution (from supply chain adjustments to real-time financial forecasting) the traditional “Operator CEO” is obsolete. The primary risk today isn’t moving too slowly; it’s moving with “automated certainty” toward the wrong horizon.

To survive, leaders must escape the Execution Trap: the seductive belief that because the data is fast and the output is polished, the direction is correct. True leadership has moved upstream. Success now depends on Judgment Salience.

The ability to apply human context, ethical friction, and “counter-intuitive intuition” to machine-optimised paths.

The 2026 CEO must pivot from managing ‘how’ work is done to governing ‘why’ and ‘if’ it should be done. As AI commoditizes execution, your only non-replicable asset is high-stakes human judgment.

Why does your “Data-First” strategy feel like a blind alley?

Most CEOs are currently suffering from Cognitive Deference. The tendency to over-weight AI-generated strategic paths because they are presented with “synthetic clarity.”

When your dashboard shows a 92% probability of success for a market pivot, the pressure to execute is immense. However, AI models optimize for historical patterns, not black swan disruptions.

If you are only reviewing the “how” of the execution, you are acting as a glorified project manager.

You must reintroduce “Human Friction” into the decision loop to stress-test assumptions that data cannot see, such as geopolitical nuance or the “soul” of your brand’s reputation.

Data-driven strategies often fail in 2026 because they lack “Human Friction,” leading to a loss of strategic agency. CEOs must move from validating AI outputs to challenging the underlying logic of machine-optimized paths.

How do you reclaim agency using “The Velocity-Vulnerability Paradox”?

The most successful CEOs I coach in 2026 utilize a framework I call The Velocity-Vulnerability Paradox.

It suggests that the faster your organization’s execution velocity becomes (thanks to AI), the more strategically vulnerable you are to small, unexamined errors.

To counter this, we implement three “Discerner Rituals”:

1. The Pre-Mortem of the ‘Perfect’ Plan: Specifically identifying where the AI’s “clean” logic ignores “messy” human variables.

2. The 10% Intuition Tax: Reserving 10% of every major pivot for “gut-check” adjustments that contradict the data.

3. The Ambiguity Hold: Deliberately delaying a “perfectly optimized” decision by 24 hours to allow for subconscious synthesis—a process AI cannot replicate.

The Velocity-Vulnerability Paradox states that high execution speed increases the impact of unexamined logic errors. Reclaim agency by implementing “Discerner Rituals” that prioritize human intuition and deliberate reflection over automated speed.

Why does an “Open Door” policy actually kill your team’s 2026 productivity?

In a world of constant Slack/Huddle pings and “AI-summarized” meetings, the “Open Door” has become a “Digital Floodgate.”

For a CEO, being “accessible” often means you are the ultimate bottleneck for low-level decisions that agents should have handled.

In 2026, your team doesn’t need your access; they need your absence to develop their own judgment muscles.

True “Human-Centered Leadership” today means protecting your “Deep Thinking” time so that when you do speak, your judgment is sharp, rested, and high-leverage.

Traditional “Open Door” policies lead to decision bottlenecks and executive burnout in AI-driven environments. Modern CEOs must protect their “Thinking Time” to maintain the high-leverage judgment required for C-suite leadership.

In 2026, the primary threat to a scaling CEO is no longer a lack of speed, but “Operational Interest.” As leadership teams increasingly delegate complex workflows to autonomous AI agents, they are inadvertently accruing a new form of “Executive Debt.” This is the distance between a CEO’s strategic intent and the actual, micro-decisions made by agentic systems. To solve this, CEOs must shift from “Delegation” to “Sovereignty Mapping.” By implementing the Velocity-Vulnerability Paradox framework, a leader ensures that increased operational speed (Velocity) does not lead to a catastrophic loss of institutional truth (Vulnerability).

Why does “Empowering” your team with AI agents often lead to executive paralysis?

For decades, the gold standard of leadership was to hire great people and get out of their way. In 2026, that mantra is a trap. When you “empower” a VP who then “empowers” a suite of autonomous agents, you aren’t just delegating - you are distancing.

Traditional delegation assumes a human feedback loop where nuance is preserved. Agentic delegation, however, optimizes for the metric provided, often at the expense of the mission intended. The result is a “drift” where the company moves at 10x speed toward a destination the CEO never chose. By the time the drift is visible in the quarterly data, the “Operational Interest” has become too expensive to pay off without a total organizational reset.

Empowerment without “Logic Sovereignty” creates a feedback loop where AI agents optimize for narrow metrics, causing a silent strategic drift that eventually paralyzes the CEO with “Operational Interest.”

What is the “Velocity-Vulnerability Paradox” in modern scaling?

The Velocity-Vulnerability Paradox, posits that the more you automate the execution of your vision, the more vulnerable your vision becomes to corruption. In a high-velocity environment, your “Sovereignty” is the only thing that doesn’t scale - and that’s exactly why it’s your most valuable asset. The paradox works like this:

1. Velocity: The speed at which your organisation processes information and executes tasks via human-AI hybrids.

2. Vulnerability: The degree to which your core values and strategic “DNA” can be diluted by automated decision-making.

As Velocity increases, Vulnerability rises exponentially unless you implement “Friction-Flashpoints” - deliberate, high-stakes human interventions where the CEO’s unique judgment is the only authorised key.

The Velocity-Vulnerability Paradox dictates that as execution speed increases, the risk of strategic dilution grows; leaders must counter this by embedding non-negotiable human “Friction-Flashpoints” in automated workflows.

How do you spot “Operational Interest” before it bankrupts your roadmap?

“Operational Interest” is the compound cost of every micro-decision made by an agent or a mid-level manager that wasn’t perfectly aligned with your 3-year “North Star.” You can spot it through three specific symptoms:

• The Consensus Mirage: Your team is in total agreement, but the actual market results are “uncannily” off-target.

• The Nuance Tax: You find yourself spending 80% of your time correcting “small” errors that have massive downstream consequences.

• The Agentic Echo Chamber: Your AI-driven reporting tools are telling you exactly what you want to hear because they’ve optimized for “CEO Satisfaction” rather than “Objective Reality.”

Operational Interest is identified by the Consensus Mirage and the Nuance Tax; it occurs when a CEO’s time is consumed by correcting “micro-drifts” rather than steering the “macro-vision.”

Why does an “Open Door Policy” actually kill productivity in an AI-augmented company?

In the past, an open door meant accessibility. Today, it means you are an un-indexed data source for everyone else’s agents. If your team can “ping” you for a quick decision at any moment, they will never build the “Decision Logic” required for true autonomy.

An “Open Door” in 2026 is an invitation for your team to outsource their courage to you. Sovereignty requires you to close the door and instead provide the “Sovereignty Ledger” - a living document of the principles behind your decisions, allowing agents and humans alike to “calculate” your response without needing your presence.

Modern Open Door policies turn CEOs into high-latency bottlenecks for their team’s AI agents; replace accessibility with a “Sovereignty Ledger” that scales your decision-making logic asynchronously.

There is a conversation that happens in almost every growing company, usually somewhere between 40 and 70 people. Someone on the leadership team says, “We need to protect the culture.” It sounds right. It feels urgent. And it leads to a series of well-intentioned efforts that mostly do not work.

Posters. A rewritten values page. An offsite that produces five words meant to define who you are. These things feel productive in the moment and are functionally irrelevant within a fortnight.

The problem is not that leaders care about culture. The problem is that “scaling culture” is the wrong frame entirely.

The 50-Person Cliff

At about 50 people, something breaks. Not dramatically - more like a slow leak. When you had 20 people, culture was ambient. It lived in how the founders spoke in standups, how someone pushed back in a product review, the particular way a first head of sales handled an unhappy client. Nobody had to explain it. You absorbed it by proximity.

Past 50, proximity disappears. New hires join teams led by managers who joined six months ago, who were themselves onboarded by someone who had been there barely a year. The original signal degrades with every relay.

Most companies respond by trying to codify the feeling. Turn it into a programme. Run training sessions. What they are really doing is trying to bottle a set of instincts that developed organically, then distribute them through slides and workshops to people who were not there when those instincts formed. It is, at best, a description of something that used to be true.

Culture Is a Decision Pattern

Here is what gets missed. Culture is not a feeling. It is not a set of words on a page. Culture is the pattern of decisions your people make when you are not in the room.

When a customer success manager decides whether to issue a refund without asking permission. When an engineer chooses to ship something 80% done or hold it another sprint. When a hiring manager passes on a strong candidate because something felt off in the values conversation - or, more revealingly, when they do not pass because the role has been open for three months and the pressure is unbearable.

Those micro-decisions, hundreds of them every week, are your actual culture. Not your values poster.

So the question should not be “how do we scale our culture?” It should be: how do we scale the criteria people use to make decisions?

Values Are Not Criteria

This distinction matters more than it first appears. Values are abstract. “We put the customer first” is a value. A decision-making criterion is something closer to: “If a fix costs less than 500 pounds and the customer has been waiting more than 48 hours, any team lead can authorise it without escalation.”

That is specific enough to act on. Specific enough that two different people, in two different offices, on two different continents, would make roughly the same call.

Without that specificity, you get the pattern every growing company recognises: same stated values, wildly different outcomes across teams. Two warehouse managers, same “ownership” culture, completely different approaches to returns, overtime, damaged stock. The value was shared. The interpretation was not. That gap is where culture actually fails, and no amount of restatement fixes it.

The other distinction worth making: a script tells someone what to do. Criteria tell them how to think. Scripts create dependency. Criteria create people capable of handling situations you have not anticipated yet, which - let us be practical - is most of them.

Where to Start

Start with the decisions that go wrong most often. Not in theory. Actually look at the last quarter. Where did someone make a call that surprised you? Where did two managers in similar situations arrive at opposite conclusions? Where did something escalate that should not have, or the reverse?

These pressure points typically cluster around five areas. Resource allocation when priorities compete. Customer exceptions and when to say yes versus hold the line. Hiring bar, specifically what “good enough” means when a role has been open 90 days. Speed versus quality trade-offs on product. And how much autonomy a team lead has before they need sign-off.

For each area, write down - not in a manifesto, in a paragraph or two - what you would weigh, what you would optimise for, where the line is. Then pressure-test those criteria with your managers. Not in a town hall. In small conversations with real scenarios pulled from the past six months. If they come back with the same answer you would give, the criteria are clear. If they come back with something that makes you wince, they are not - or you have not actually communicated them.

That gap between intended criteria and interpreted criteria is your actual culture problem, made measurable. Most organisations have alignment on maybe a third of their critical decision types. The rest is guesswork dressed up as autonomy.

The Counterintuitive Part

Clear decision criteria do not produce uniformity. Or rather - wait, they do produce uniformity on the things that matter, but they also produce more confidence and, oddly, more variation on everything else. People who know where the boundaries are tend to be more creative within them. People who are guessing at the boundaries play it safe or do whatever they feel like, neither of which is useful.

The data on this is fairly consistent. Organisations that move from vague values to specific decision frameworks typically see escalation rates drop by half or more within a quarter. Customer-facing teams make faster calls. And satisfaction scores tend to go up, not down, because a decision made quickly by someone who knows the framework tends to feel better than a slow decision made by someone checking with their boss.

There is an uncomfortable side effect, though. Making criteria explicit sometimes surfaces genuine disagreements that leadership has been papering over with vague language. Two co-founders who discover they have been optimising for different things for two years. A leadership team that finds its stated priorities actually contradict each other when applied to real trade-offs. This is painful but probably overdue. Ambiguity is not alignment. It is just deferred conflict.

Why This Becomes Self-Reinforcing

When you scale decision-making criteria rather than culture slogans, the system starts to maintain itself. New managers learn the criteria through real decisions, not workshops. They pass them on the same way. When someone makes a call that does not fit, it stands out - not because it violated something vague, but because it deviated from a framework everyone else is using.

You stop needing to be the culture police. The criteria do that work. Not perfectly - there will always be edge cases, and you will need to update things as the business changes. But the baseline holds without you personally reinforcing it in every all-hands.

Compare that to the alternative: trying to keep a feeling alive across hundreds or thousands of people through storytelling and offsites and repetition. One of those scales. The other requires you to be everywhere at once, which past 50 people you are not and never will be again.

The Real Shift

This is not about reducing culture to a set of rules. The criteria are meant to carry the same instincts that made your culture work when it was small. They are a translation mechanism, not a replacement. The goal is that when someone three layers removed from you faces a hard call at 6pm on a Friday, they make roughly the same decision you would. Not because they have memorised your values page, but because they understand how you think about trade-offs.

That is what culture at scale actually looks like. Not a poster. Not a programme. A shared way of making decisions that works whether you are in the room or not.

Which, past 50 people, you mostly are not.

Here is something most CEOs already know but rarely explicitly acknowledge. The strategy is fine. The offsite produced something reasonable. The board liked it. The leadership team nodded. And then it entered the organisation and quietly starved to death.

Not because anyone disagreed with it. Not because the market shifted overnight. It starved because nobody moved anything out of the way to feed it. The calendar - the actual week-by-week, meeting-by-meeting calendar of your senior leaders - kept running on last quarter’s logic. And strategy, it turns out, cannot survive on enthusiasm alone. It needs hours.

I have sat in more post-offsite debriefs than I would like to admit where a CEO asks, genuinely frustrated, why a strategic priority stalled. The answer, almostalways, is sitting in Outlook. Or Google Calendar. Or whatever scheduling tool the organisation uses. Pull up the calendars of the eight or ten people who were supposed to carry this thing forward. Look at what actually got their time in the six weeks after the offsite. I would wager fewer than 20% of their hours connected to the work that was supposedly the most important thing the company needed to do.

That is not a communication problem. That is a physics problem.

The Calendar Never Lies

There is a useful exercise I sometimes recommend to leadership teams, though fair warning, it tends to make people uncomfortable. Take your top three strategic priorities - the ones you put on slides, the ones you told the board about - and then audit your senior leaders’ calendars against them. Not what they say they are working on. What they are actually spending time on, meeting by meeting, week by week.

The gap is typically brutal. One CEO I worked with found that her VP of Product was spending roughly 60% of his week in recurring meetings that predated the strategy by two years.

Operational reviews, cross-functional syncs, pipeline updates - all perfectly reasonable in isolation, all completely disconnected from the three things the company said it needed to do in the next twelve months. He was busy. He was visibly busy. But he was busy doing 2024’s job in 2026.

And this is where it gets awkward, because the VP was not being negligent. He was doing exactly what the organisation’s operating rhythm told him to do.

Nobody had cancelled anything. Nobody had said, “These six hours of weekly meetings? We are stopping three of them because the world changed.” The old calendar just... persisted. Strategy was an addition. Everything else was an inheritance.

Recurring Meetings Are Organisational Cholesterol

I want to be specific about what is happening here because the instinct is to frame this as a time management issue, and it is not. This is a structural problem with how organisations allocate their most constrained resource. A senior leader typically has about 40 to 45 hours of working time per week. By the time you subtract the meetings that were already recurring before any new strategy was announced - operational reviews, team standups, one-on-ones, leadership syncs,

cross-functional updates, the monthly business review that somehow takes a full afternoon - you are left with maybe eight to twelve hours of discretionary time. On a good week.

Now ask that person to lead a major strategic initiative. Where does the time come from?

Usually it comes from nowhere. The initiative gets slotted into the margins. A 30-minute weekly check-in gets created - another recurring meeting, ironically - and everyone agrees to “move fast” on it. But moving fast on something you can only think about for four fragmented hours a week is not really moving fast. It is moving at the speed of whatever time is left over after the old operating model takes its cut.

Recurring meetings are the real operating system of most companies. They define what gets attention, what gets debated, what gets decided. And they are extraordinarily sticky.

I have seen recurring meetings survive three reorganisations and two changes of leadership. Nobody remembers why the Thursday afternoon pipeline review exists, but it is in the calendar and so it continues to happen, consuming 90 minutes of six senior people’s time every single week.

That is nine person-hours. Roughly 400 person-hours a year. For a meeting no one can explain the purpose of.The truth is someone probably can explain the purpose. They will say it is for “alignment” or “visibility.” And they are not wrong, exactly. These meetings do provide those things, in the way that leaving all your lights on provides illumination. The question is whether the cost is proportionate and whether anyone is checking.

Busy on Strategy Versus Busy Near Strategy

There is a distinction I find useful, though I should say upfront that it irritates people when you first introduce it. The distinction is between being busy on strategy and being busy near strategy.

Most senior leaders are doing the second and reporting the first.

Being busy near strategy means your work is adjacent to the priority. You are in meetings where the strategic initiative gets mentioned. You are cc’d on the emails. You have opinions when asked. But your actual output - the decisions you are making, the problems you are solving, the things that keep you up at night - those are still anchored to the previous operating model. You are orbiting the strategy without ever landing on it.

Being busy on strategy looks different. It means the initiative is consuming your best hours, not your leftover ones. It means you have cancelled or delegated something else to make room. It means the work shows up in your calendar as blocks of protected time, not as a recurring 30-minute sync that keeps getting bumped for “something urgent.”

Most organisations have no mechanism for distinguishing between these two states. When the CEO asks for a status update, they hear that people are “working on it” and “making progress” - which is technically true, in the same way that walking slowly in the direction of a destination counts as making progress.

The question nobody asks is: how many hours this week did you actually spend on this? And if the answer is honest, it is usually something like two. Maybe three.

The “Just This Once” Problem

Even when leaders do protect time for strategic work, it has a half-life of about two weeks. Someone protects Friday mornings for deep work on the new market entry. Week one, it holds. Week two, a client escalation eats into it - “just this once.” Week three, a direct report needs the slot for a conversation that feels urgent. By week four, the protected time is gone and nobody notices because nobody was tracking it.

This is where the problem becomes structural and not behavioural. You cannot solve it by telling people to be more disciplined. The forces acting against protected strategic time are persistent and legitimate - real clients, real crises, real team needs.

The issue is that strategy has no such constituency.

Nobody is going to send an urgent Slack message saying, “The strategic initiative is being neglected and needs your attention right now.” Operational work has urgency built into it. Strategic work does not. It is always the thing that can wait until next week.

And so it waits. And waits. And six months later someone puts together a slide deck asking what went wrong with the initiative, and the post-mortem focuses on market conditions or resource constraints or cross-functional misalignment.

Nobody writes “we never actually gave it any time” because that sounds too simple.

Too embarrassing. But it is frequently the actual answer.

What a CEO Can Actually Do About This

The obvious response is to tell people to clear their calendars, and the obvious response is wrong. Or - let me qualify that - it is incomplete. Telling senior leaders to make time for strategy without removing anything from their plate is just adding pressure.

What gets results, in my experience, is something more uncomfortable: a subtraction exercise before the addition.

One approach that I have seen work, and it takes more nerve than you might expect, is to run a calendar audit at the leadership team level. Not as a productivity exercise. As a strategic resource allocation review. You take the top three priorities the company committed to. You pull calendars for the ten most senior people involved. And you map, honestly, where their time went over the past four to six weeks.

The first time you do this, expect defensiveness.

People will explain why their meetings matter. And some of those explanations will be valid. The point is not to shame anyone. The point is to make the invisible trade-off visible. Every hour spent in a recurring operational meeting is an hour not spent on the thing you told the board was your top priority. That is a real cost, and most organisations are not accounting for it.

After the audit, the harder work begins. You have to kill things. Specific meetings, specific processes, specific reporting cadences. Not all of them - some operational rhythms are genuinely necessary. But in most organisations I have worked with, something like 25-30% of recurring senior leadership time is going to meetings that serve a purpose that could be handled asynchronously, delegated to someone else, or simply stopped.

A technology company I advised in 2023 went through this exercise and freed up roughly eleven hours per week across their eight-person leadership team. That is not a marginal gain. That is the equivalent of hiring a senior person and telling them to work on nothing but the strategy. Except it cost nothing and happened in two weeks.

Making the Calendar the Accountability Mechanism

The second shift, and this is where it gets genuinely countercultural for most organisations, is to treat calendar allocation as a leading indicator of strategic execution. Not the quarterly business review. Not the OKR check-in. The calendar.

Think about it this way. If your CFO told you that 80% of the budget was committed to maintaining existing operations and only 12% was allocated to the new strategic priorities, you would be concerned. You would probably restructure the budget. But when 80% of your leadership team’s time is committed to maintaining existing operations and 12% goes to the strategy, that somehow feels normal.

It is the same resource allocation problem. We just do not treat time the way we treat money.

Some organisations have started doing bi-weekly or monthly time audits alongside their financial reviews. It is simple, bordering on crude: what percentage of your senior leaders’ time went to strategic priorities versus operational maintenance? The target does not need to be 50-50. For most companies, something like 25-30% of leadership time on strategic work is ambitious enough. The point is that you are measuring it. You are making it visible. You are creating a conversation about it that happens regularly, not just when something has already failed.

The Uncomfortable Truth

I realise this entire argument can sound reductive. You spent a quarter of a million on a strategy consultancy and the answer is to look at people’s calendars? Well, sort of. The strategy itself might be excellent. The analysis might be rigorous. The priorities might be exactly right. None of that matters if the organisation does not rearrange its time to match.

There is a version of this argument that blames culture, or middle management, or change resistance. And those things are real. But they are often symptoms of something more basic: the operating rhythm did not change. The meetings did not change. The weekly cadence did not change. People were told to do something new without being told to stop doing something old. And given a choice between the meeting that already exists in their calendar and the initiative that requires them to create new habits from scratch... well, inertia wins almost every time.

The calendar is not a metaphor for your strategy. It is your strategy. Or, more precisely, it is the strategy that your organisation is actually executing, regardless of what the slides say. And if those two things do not match - if the calendar tells a different story than the offsite deck - then you do not have an execution problem. You have a resource allocation problem hiding in plain sight.

The fix is not complicated. It is just uncomfortable. Because it means looking at how your most senior people actually spend their time, and admitting that a lot of it is not serving the thing you said matters most. That conversation - specific, data-driven, slightly painful - is where strategy execution actually starts.

Everything before that is aspiration.

If your “Strategy 2026” document is a 50-page slide deck filled with market analysis, competitive landscapes, and 14 different “key initiatives,” you haven’t written a strategy. You’ve written a security blanket. You are planning for every possible outcome because you are too afraid to bet on one.

Planning is an addition problem. You look at your resources, your team, and your market, and you ask: “What else can we do?” It feels productive. It keeps the board happy because the slides look professional and the spreadsheets balance.

But Strategy is a subtraction problem. A real strategy is a clinical, often painful set of choices that define what you will not do. If your strategy doesn’t make at least three department heads or investors uncomfortable, it isn’t a strategy - it’s a consensus. And consensus is where high-growth companies go to die.

To move from a bloated plan to a high-signal strategy, run your current “Strategic Initiatives” through the Subtraction Test:

1. The Resource Collision: If you had to double the budget and headcount for your #1 priority tomorrow, which three “important” projects would you kill to fund it?

2. The “No” Audit: Review your last three Board decks. List every opportunity you turned down. If that list is empty, you don’t have a strategy; you have an “Open Door” policy for distractions.

3. The Front-Line Translation: If you ask a junior developer or a sales rep what the strategy is, can they answer in one sentence? If they need a slide deck to explain it, the strategy is too complex to execute.

Kill the fluff. Take your current list of “Top 10 Priorities” and apply the Rule of Three.

• Identify the top 3 that actually move the needle.

• Take the remaining 7 and move them to a “Not Now” list.

• Communicate to your team that any work spent on those 7 is officially a distraction.

Strategy isn’t about being right; it’s about being focused. You can’t win a war by fighting on every front.

There’s a particular kind of resignation letter that HR directors recognise immediately. Not the angry one, not the one citing a competitor’s salary package. The quiet one. Two paragraphs. Vague language about “new opportunities.” Written by someone who stopped caring about six months ago and has simply been waiting for the right moment to make it official.

That letter is not a decision. It’s a conclusion.

The decision happened earlier - maybe at a Monday morning meeting where someone else got the credit for a project they barely touched. Maybe when a piece of feedback they raised three times finally got actioned, but attributed upward rather than sideways. Maybe it was nothing so specific. Maybe it was just the slow accumulation of being treated like a resource rather than a person, until eventually the mental maths stopped adding up.

I’ve sat in enough exit interviews to know that people rarely say the real reason out loud. They’re professional. They’re already mentally gone. They mention “career development” or “a change of direction” because it’s easier than explaining that their manager never once asked how they were actually doing.

Here’s what I think CEOs consistently underestimate. The cost isn’t the recruitment fee. That’s visible, it goes on a spreadsheet, someone signs off on it. The cost is the eighteen months before the letter. The coasting. The institutional knowledge quietly walking out the door to competitors. The team members watching it happen and updating their own CVs accordingly.

One departure rarely travels alone.

There’s decent research suggesting that replacing a mid-level employee costs somewhere between 50 and 200 percent of their annual salary, depending on the role. Most organisations I’ve encountered don’t track that figure with any rigour - which is convenient, because if they did, the business case for basic human decency would be almost embarrassingly obvious.

And yet. Leaders who would scrutinise a £20,000 procurement decision will somehow never find five minutes to tell someone they did good work.

The signals are usually there. You’ve seen them yourself, probably. The person who used to arrive early and started leaving on time. The one who stopped offering ideas in meetings - not because the ideas dried up, but because the ideas kept getting absorbed and redistributed without acknowledgement. The shift is gradual enough that you can explain it away each time. Busy week. Bit tired. Going through something personal, maybe.

I’m not saying you’re deliberately ignoring it. Most managers aren’t. They’re busy, they’re managing upward pressure themselves, they’re dealing with the operational noise that fills every working week. The failure is usually one of attention, not intention. Which is a meaningful distinction, but it doesn’t change what happens next.

Once someone has made the internal decision to leave - even if they haven’t told anyone, even if they’re not sure themselves yet - the dynamic shifts in ways that compound. They stop going the extra mile on things that won’t be on their CV. They disengage from the longer-horizon work. They protect their energy. Entirely rational behaviour, all of it. And almost impossible to reverse without a significant intervention that most organisations never quite get around to.

The uncomfortable question for any leadership team is this: what are we actually reinforcing?

Because in a lot of organisations, the behaviour that gets rewarded isn’t the quiet person doing excellent, unglamorous work over a sustained period. It’s the visible person. The one who manages up well, who’s in the room when decisions get made, who knows how to position their contributions. That’s not entirely wrong - communication and visibility matter in organisational life. But when the gap between who’s genuinely delivering and who’s getting the credit becomes too obvious, people notice. The ones doing the unglamorous work notice most of all.

Loyalty - real loyalty, the kind where someone turns down a better offer because they believe in what they’re building - typically comes from feeling that the exchange is roughly fair. Not perfectly fair. Nobody expects that. But fair enough. Seen. Valued at something approximating what they’re actually worth.

The practical things matter more than most senior leaders want to admit. Following through on commitments. Saying thank you in a way that sounds like you mean it rather than like you’re ticking a box. Giving credit publicly, specifically, to the person who actually did the work. These aren’t complicated. They’re also, mysteriously, inconsistently done.

None of this is a revelation. Most leaders, if you asked them directly, would tell you that people are their most important asset - and would be embarrassed to be caught not believing it. The problem isn’t knowledge. It’s the gap between knowing something and building it into the actual texture of how you lead week to week.

Retention strategies as a formal exercise tend to miss the point slightly. The perks, the engagement surveys, the quarterly town halls where the CEO is very relatable for forty minutes. None of that lands if the daily experience of working in your organisation is one of being taken for granted. People can hold two ideas simultaneously: the company says it values them, and their day-to-day reality suggests otherwise. The second one wins. It always wins.

The leaders I’ve seen retain genuinely exceptional people over time don’t have secret techniques. They’re just consistently attentive. They notice when someone’s done something difficult. They remember what people said they cared about. They make the small corrections early, before the quiet exit has already begun.

By the time someone’s protecting their peace rather than trying to prove their worth - to use a phrase I’ve heard a few times recently - you’re not in a retention conversation anymore. You’re in a departure conversation that hasn’t been announced yet.

Start earlier. It’s cheaper. And it turns out, it’s also just better.

There’s a meeting I keep seeing inside companies. It happens in Q3 usually, when the numbers are ambiguous and the board is restless. A senior team sits around a table, or a screen, and someone says: “We need more data before we can decide this.” And heads nod. It feels prudent. It sounds like good governance.

What it typically is, though, is a way to avoid the decision for another six weeks.

This isn’t a simple story about indecision masquerading as rigour. Some decisions genuinely do need more information. A pharmaceutical firm deciding whether to proceed to Phase 3 trials is not the same as a retail brand deciding whether to exit a mid-tier city market. The stakes, the reversibility, the cost of being wrong - these things matter. So I’m not going to tell you to trust your gut and move fast. That’s its own kind of laziness.

But there is something real happening inside organisations that collect data compulsively, commission research reflexively, and still find themselves unable to act. And it’s worth naming.

The problem isn’t that people are stupid

The executives I’ve worked with who are slowest to decide are often the most intelligent people in the room. That’s not incidental. They can see more angles. They hold more of the uncertainty in their heads at once. They know what they don’t know, which is more than most people can say.

The problem is that information, past a certain point, doesn’t reduce uncertainty. It redistributes it. You find out one thing and it opens three new questions. You commission a customer survey and the results are directional but the sample feels thin so you do another one. You get the second one back and it mostly confirms the first but there’s one segment behaving oddly and now you need to understand that segment. And so on.

It becomes a kind of archaeology. You keep digging not because you expect to find the answer at the next layer but because stopping feels arbitrary.

What the information is actually doing

Here is something I’ve found to be true more often than I expected: in most cases, the senior team already knows what they need to do. Not always, but often. The decision has been quietly visible for months - in the churn numbers, in the conversations the CFO is having with the CEO on Tuesday evenings, in the fact that the same three people keep being careful about how they phrase things in public forums.

The information gathering isn’t filling a gap in knowledge. It’s performing the process that allows everyone to say, later, that they were responsible stewards of the decision. That it was evidence-based. That no one rushed.

There’s nothing necessarily wrong about this. Process legitimacy matters. You can’t run a large organisation purely on instinct, or on the read of one senior person who happens to be right this time. But the performance becomes a problem when it delays action past the point where action can actually help, or when it gives cover to a faction within the leadership team who want the decision to not happen at all.

Wait - let me explain that last part more precisely. I’ve seen situations where one or two members of a leadership team are implicitly opposed to a course of action but lack the standing, or the appetite, to say so directly. What they can do is request more information. Call for another round of analysis. Suggest that the market picture isn’t yet clear enough. This is almost impossible to argue against in real time because it sounds responsible. But it functions as a veto dressed up as diligence.

The 70% rule, and why it’s harder than it sounds

Jeff Bezos wrote something in his 2015 shareholder letter that gets cited a lot in these conversations: the idea that most decisions should be made with roughly 70% of the information you wish you had. Waiting for more, he argued, is probably too late most of the time.

This is roughly correct, in my experience. The problem is that 70% is genuinely hard to identify when you’re inside the decision. You don’t know what the remaining 30% would tell you. It might be nothing. It might be something that changes the whole picture. That’s precisely the feeling that makes people want to go and find it.

The way I’ve seen executives navigate this reasonably well is by separating two distinct questions that tend to get conflated in these meetings. First: what would change our decision? Second: can we actually get that information in a useful timeframe?

If the answer to the second question is no - if the information is six months away, or methodologically impossible, or contingent on things outside your control - then its absence shouldn’t delay you. You’re not going to get it. You’re deciding in the dark, same as you would be later, except later you’ll have less runway.

If the answer is yes - if the information is genuinely obtainable in two weeks and would materially shift the decision - then fine, go get it. But put a boundary on it. Decide on Thursday. The research comes back or it doesn’t, and either way you decide on Thursday.

The data volume problem is different from the data quality problem

These two things get muddled. Companies sometimes believe they have an information problem when what they actually have is an interpretation problem.

I spent time with a business in 2024 - mid-sized, operating across 20 countries. They had accumulated a remarkable amount of operational data. A genuinely impressive dataset. And yet they kept commissioning more research because the executive team couldn’t agree on what the existing data was saying.

That’s not an information problem. That’s a disagreement about assumptions, dressed up as a gap in evidence. No amount of additional data was going to resolve it, because the underlying question was about strategic priorities, not facts. They needed a different kind of conversation.

More data in that context was actually making things worse. It was providing more surface area for disagreement. More numbers that could be read in different ways by people who had different underlying views about what the business was for.

The cost nobody puts in the spreadsheet

Slow decisions have costs that are easy to undercount because they’re largely invisible.

The competitor who moved six months earlier and is now embedded with the customer you were going to approach. The talented person who left because they couldn’t see the company committing to anything. The organisational energy - and I mean this quite literally, the actual psychic energy of the people inside the business - spent in holding patterns, waiting, preparing updated analyses for meetings that conclude with the formation of a workstream.

That last one is underrated. People inside organisations can feel when the information-gathering has become chronic. They learn to stop expecting decisions to come out of decision-making meetings. They start to front-run the outcome by hedging in their own work. And slowly the culture shifts from one where initiative is valued to one where caution is the only rational strategy.

This is typically irreversible once it’s embedded. You can change a decision-making process in a week. Changing a decision-making culture takes years.

What faster decisions actually require

This is not primarily a process problem, which is why process fixes typically don’t work for long.

It’s a risk tolerance problem, and specifically, a tolerance for personal accountability. Fast decisions in uncertain environments mean that sometimes you’re going to be visibly, attributably wrong. The decision had your name on it. The board knows. The team knows.

In organisations where failure is expensive - not just financially, but in terms of status, career, relationships - people rationally protect themselves by diffusing accountability. More stakeholders consulted, more data gathered, more consensus built. If it goes wrong, it went wrong collectively. The information-gathering process is partly how you arrange that.

So if you want faster decisions without just demanding that people take more personal risk - which is not really a workable ask - you have to change what happens when someone makes a call and it doesn’t pan out. That means something different in practice in every organisation, but it usually involves senior leaders being genuinely, non-performatively public about their own misjudgements. Not in a “fail fast” culture-deck kind of way. In an actual, specific, “I thought we’d hold market share in that segment and we didn’t, here’s what I misread” kind of way.

That’s uncomfortable. Most senior leaders avoid it. And that avoidance, more than almost anything else, is what produces the excess-information culture that slows the whole enterprise down.

A different posture

The executives I’ve seen handle this well tend to carry a specific habit: they’re clear, in their own minds, about what kind of decision they’re making before they ask for information.

Is this a one-way door or a two-way door? How much would it cost to reverse course if we’re wrong? What’s the decision actually contingent on - customer behaviour, competitive response, internal capability? Of those contingencies, which ones do we genuinely not know, versus which ones are we just uncertain about?

That last distinction matters a lot. Uncertainty you can act under. Genuine ignorance - where you’re missing a fact that you could reasonably obtain - is different.

Most decisions, honestly, land in the uncertainty category. You know more or less what’s likely. You just don’t know for certain. And certainty, in most markets, in most competitive environments, is simply not available. Waiting for it is waiting for something that isn’t coming.

The question is never really “do we have enough information?” The question is “what are we actually waiting for, and does waiting actually help us?”

Ask that in your next meeting where someone suggests commissioning further research. The answer will tell you quite a lot.

You look at a modern boardroom now, and honestly, you see a bit of a challenge. The old ways of doing things, the usual governance rhythm, it’s struggling. The sheer speed of the digital world, the generative AI stuff that’s everywhere, the constant fear of a cyber-attack, plus all that technical debt we’ve built up over years... it means the Non-Executive Director role has to change.

It’s not enough to be a sensible generalist anymore, is it? We need someone who genuinely gets the digital strategy. I mean, the necessity for a Digital NED isn’t just a trend. It’s fundamental for keeping the organisation on the rails, and ensuring long-term resilience.

If technology is the main business strategy now, then the board’s job, their whole ability to ask the right questions about risk, ethics, and value, hinges entirely on having that specialised digital insight around the table.

A Technology Gap in the Boardroom

I’ve found that there’s this really deep technology deficit undermining boards. It’s pervasive. The research I’ve conducted is pretty stark. There’s this enormous gulf between the digital engagement a company needs and what you actually find in the boardroom. It’s a structural issue, not just a lack of technical lingo.

Business leaders typically flag this deficit because it just hampers growth. You might find boards are deeply uncomfortable trying to weigh up how digital transformation is actually going. When the board can’t offer proper stewardship, it can actually become the thing that stops the necessary changes from happening. Funny, that.

The friction between the board and the executive team is a clear symptom. When I speak to people, roughly 40% of directors admit they rely too much on management for tech insights. At the same time, C-suite people feel just as strongly. 44% of them are concerned their board simply lacks the fluency. This creates a really messy dynamic. Tech leaders feel unheard. Board members feel overwhelmed. When you see that only 36% of board members are expressing full confidence in their technology leadership teams, you realise the organisation’s ability to pull off complex, multiyear changes is just severely compromised.

And let’s talk about money for a minute. The inability to clearly link tech spending to actual business outcomes is maybe the most damaging consequence of this gap.

Over 42% of both executives and directors struggle to make that link between investment and growth. You also get around 25% pointing to fragmented reporting, inconsistent KPIs;. These are real barriers to showing an actual return on investment, or ROI. It leads boards to focus on short-term, instant ROI-driven thinking, not long-term value.

Without a Digital NED there to challenge those narrow measures, a company might easily under-invest in the very technologies, like, say, better cloud infrastructure or advanced analytics, that they need to keep ahead of the competition and grab new opportunities.

What is a Digital NED, Actually?

The Non-Executive Director’s role is classic. Independent oversight, challenging the executive team, bringing external experience. But in the digital age, that role now demands what people call a “T-shaped” profile. That means serious business gravitas combined with digital incisiveness. A Digital NED, it’s important to remember, is not an IT technician. They’re a senior leader who understands how technology fundamentally changes market behaviour, customer habits, and the risk profile of the organisation.

Their real value is in bridging that gap, translating technical complexity into strategic clarity. Making sure the substance is what matters in the meeting, not just status.

You see, effective Digital NEDs need a mix of hard and soft skills to be credible. Sure, they should grasp the basics of cybersecurity protocols, data laws, and the AI framework landscape. But they also have to be long-term thinkers, active listeners, and be able to challenge management without being destructive. They’re independent in that they don’t have the day-to-day operational bias of the executive directors. They have formal board responsibilities too, which is a key differentiator - full fiduciary duties and legal obligations to the company and its shareholders.

In my experience, when a startup or a scaleup appoints their first Digital NED, it usually signals a jump in maturity. They are moving away from that single, founder-led decision-making. That NED provides absolutely crucial mentorship, helping those leaders separate personal risk from company risk. It gets the organisation ready for scrutiny, maybe for external investment.

They are, in essence, guardians of transparency. Identifying those blind spots that pressure can create. Bringing in perspective from other cycles, other sectors.

The Strategic Translator

One of the biggest headaches in the boardroom is the sheer language barrier between the tech people and the non-technical directors. CIOs, CTOs, CISOs, they often struggle to present their vision in a way that resonates with a board focused mainly on risk, finance, and strategy. That’s where the Digital NED acts as a “strategic translator.”

They simplify the tech concepts without losing the strategic importance. They help the board see that technology isn’t just a separate department, an overhead, but an integral part of the entire business model. It touches everything, from customer experience right through to how agile the organisation can be.

To improve these boardroom conversations, tech leaders have to stop with the tactical reports and start talking strategy. A Digital NED can push this shift, encouraging jargon-free talk. They might simplify a complex issue by using a story or a relatable example. Look, instead of spending 45 minutes on the technical specs of a new ERP system, the discussion should be focused on the why. How does it enable us to make faster decisions? How does it improve supply chain resilience? That’s what matters at that level.

A key tool I’ve seen used is the “technology scorecard.” It links those tech metrics directly to the business outcomes. You can track progress over time, even compare against competitors. Digital NEDs push for this consistent reporting structure. They’ll want to know the “Run/Grow” ratio. The amount of the IT budget spent just keeping the lights on versus what’s being invested in new, transformative initiatives.

This distinction, Run vs. Grow, is vital for spotting technical debt. Understanding the true, complex cost of that old, clunky architecture. For a board, a sample scorecard might include:

• Financial Run/Grow Ratio: That’s innovation capacity versus the maintenance burden.

• Risk Technical Debt Interest: What’s the cost of delaying system modernisation?

• Operational Time-to-Market for Digital Products: Agility, frankly. How fast are we?

• Strategic AI-Enabled Process Percentage: The honest degree of digital transformation maturity.

• Security Cyber Simulation Frequency: Organisational resilience. Are we ready?

I’d add that a good Digital NED will also foster more informal engagement. They encourage tech leaders to build rapport with board members one-on-one. Deep-dive sessions. This lets directors explore things like generative AI nuance in a way that isn’t stressful, helping the board calibrate what the organisation can actually do.

Strategy, Not Operations

There’s this constant fear among boards. Bringing in a digital expert will mean “operational drift.” They worry they’ll get too involved in the day-to-day running. But a Digital NED’s job is actually the opposite.

They make sure the board keeps a high-level, strategic focus on technology while still making sure they have enough visibility into the operational risks. Strategic and operational management, they’re separate processes, but they must both be functioning for the company to truly thrive.

Operational management is all about the what. The current flow of activities that deliver business objectives. IT infrastructure, logistics, sales support. Strategic management is about the how. That required curiosity to shape the future of the organisation, to react to market shocks. Board talks must prioritise those big strategic decisions: expanding into new markets, radical innovation. These carry high risk, high reward, and huge financial and technological investments.

To keep that focus, Digital NEDs help the board ask what I call the “do-or-die” questions. These are questions designed to really test a business plan’s robustness. To be specific, they’re meant to uncover whether the tech strategy is genuinely helping the company win against new competitors.

• Are we winning? Who are the emerging competitors, and how is our tech strategy giving us the edge over them and our traditional rivals?

• Are customers happy? How does our experience compare to the best in other sectors, and what do we need to do to exceed what they’ll expect next?

• Are we agile enough? Do our current business plans truly reflect the full potential of technology to improve how we operate?

• Can we actually do it? Have we got the skills, the data quality, and the lean processes to deliver the full value from our tech investments?

• Is our risk level right? Is our IT risk measured correctly and aligned with the company’s overall appetite? And who, specifically, is on the hook for it?

By framing technology governance this way, the board ensures strategy drives technology decisions, not the other way around. The Digital NED is a catalyst here, ensuring technology is a standard item on the agenda, not just something you talk about when there’s a massive crisis or a big budget review.

The AI Challenge (and Opportunity)

Generative AI adoption has created what some people call an “AI reckoning.” It’s not just a technical pilot anymore. It’s a strategic priority. It will either make or break a company’s advantage. I’ve seen so many companies rush into AI without a clear business case, risking “AI sprawl,” costly experiments that deliver very little ROI. A Digital NED is crucial for nailing down the organisation’s “AI posture.” It’s strategic ambition. What is its appetite for this kind of change?

Boards need to agree on which AI archetype they’re pursuing. That decision fundamentally dictates how AI is integrated and how it should be governed.

• Business Pioneer: They want to reshape the market. Focus is new revenue streams, big strategic expansion.

• Internal Transformer: They aim to rewire the operating model. Holistic integration across the entire enterprise for structural change.

• Functional Reinventor: They want to optimise specific workflows. High-value, targeted use cases within, say, Finance or HR.

• Pragmatic Adopter: They are simply following market traction. Fast-following proven capabilities just to maintain competitiveness.

Once you have that posture, ownership of oversight needs clarity. Big, material investments might go to the full board. Risk frameworks and vendor reviews might be for a committee. A core responsibility for the Digital NED is creating that AI governance policy framework. You need scaling rules for pilots, risk thresholds that absolutely require human sign-off, and IP protection rules for the training data.

Ethical literacy is the key here. Boards must move past just being technically curious. They need a deep, deep understanding of the ethical side: transparency, fairness, accountability. Directors should be asking very pointed questions, making sure AI systems are aligned with the company’s values:

• Can the company truly explain how its algorithms are making decisions? Are they auditable?

• Is the AI development following regulatory and legislative expectations? Like, say, the EU AI Act?

• How do we validate model accuracy, and how do we spot if the model starts to drift over time?

• Have we engaged “red teams” to assess our generative AI use cases for potential harms?

Ethical foresight becomes a competitive edge. Companies that prioritise good AI governance typically earn public trust and attract the best people. Those that lag will, I think, just find themselves chasing the regulators or slowly losing relevance.

Technical and Digital Debt

There’s this slow, quiet killer that affects established companies: technical debt. It’s that build-up of outdated code, old legacy systems, architectural shortcuts. It’s an anchor on innovation. It’s not just an IT problem, it’s a major strategic challenge that derails transformation and really impacts customer satisfaction. For big incumbents, an “archaeology of legacy technology” will often reveal systems that have no vendor support, are exposed to security holes, and often rely on a small team of long-serving people who are close to retirement. That’s a huge risk.

Digital debt is even broader, potentially more dangerous. It’s what happens when you accumulate digital assets, platforms, websites, processes, faster than you can optimise or retire them. This debt compounds because of departmental “shadow IT” - different teams solving immediate problems with standalone solutions that no one in the business ever strategically coordinates. The result is a fragmented customer experience, siloed data, and decision-making that is painfully slow.

You’ve got to help the board see both technical and digital debt as strategic risks.

Digital NEDs should push for technical debt audits and ensure debt reduction is explicitly part of the business roadmaps. By prioritising the debt repayment alongside the new innovation, you keep the enterprise agile and competitive. You might ask: is the maintenance spend disproportionate to the innovation investment? Is there a clear, actionable roadmap to get us to a target-state architecture?

Measuring the Return on Transformation

The ROI question around digital transformation is a constant source of boardroom frustration. Traditional financial metrics are often just not enough to assess digital initiatives, which have so many intangible benefits. Better customer loyalty, higher employee morale. McKinsey research, for instance, suggests a big gap. Big companies only capture about 31% of the expected revenue lift and 25% of the projected cost savings from digital efforts. That’s a massive shortfall.

A Digital NED ensures the board moves past simply tracking activity to measuring outcome. This needs a comprehensive framework that includes the financial impact, yes, but also operational efficiency and customer satisfaction. A major, very common, roadblock is simply the lack of baseline data.

If you don’t establish a clear starting point before you implement, you can’t accurately track progress and show the real impact.

Look at the success stories in logistics or retail. They align the technology with a specific business goal. Reducing the loan approval time, maybe, or increasing the patient bookings. Digital NEDs also highlight those hidden costs that can totally distort ROI. Things like training, the change management programme, or integrating old legacy systems. By providing this holistic view, the Digital NED makes sure digital investment actually translates into tangible, long-term business value.

Psychology and Recruitment’s Vicious Circle

Adding a Digital NED changes the boardroom dynamic. Boards are notoriously susceptible to groupthink, that desire for consensus which can easily override critical evaluation. Groupthink leads to overconfidence, less rigorous debate, and a tendency to dismiss outside advice. In high-growth firms, I’ve found that “director overlap” - when people have served together for ages - can really make groupthink worse, making the board less likely to actually innovate.

A Digital NED can mitigate this. They often act as a ‘devil’s advocate,’ injecting objective, data-driven analysis. They help the board avoid what I call “digital illiteracy,” which is when directors fall back on tired clichés like the “Kodak example” because they lack a genuine grasp of the current shifts. By encouraging constructive dissent and rewarding critical thinking, the Digital NED builds a culture where rigorous debate genuinely strengthens decision-making.

But so many boards are stuck in this recruitment “vicious circle.” They can’t define their digital strategy because they lack a digital profile on the board. But then, they can’t decide which digital profile to hire because they lack a strategy. To break it, boards need to prioritise those “T-shaped” people, and be willing to look beyond the usual suspects.

When recruiting, there’s always that dilemma: do you go for the young, Digital Native from a startup? They have the very latest technical knowledge and agile thinking, but they might lack the corporate governance experience. Or do you go for the Seasoned Corporate Leader? They have experience with large-scale transformation and the necessary boardroom gravitas, but their ideas on technology might be a bit dated. The strategy, in my experience, is to compose the board’s digital profile from both. You need that mix. Digital needs are also, thankfully, a major driver for diversity.

Regulation is Catching Up

Regulatory frameworks globally are moving quickly, putting more responsibility on boards for tech and cyber oversight. The UK Corporate Governance Code (2024 edition, effective 2025) is a big one. Boards now have to declare in their annual report that all “material controls”. Including IT and reporting controls, are operating effectively. That’s a shift from a ‘comply or explain’ approach on the process to a direct declaration of effectiveness. That means more personal accountability for directors.

In Australia, the ASX Principles (the 5th edition, 2025) are pushing the board skills matrix. Boards must disclose how they assess and then fill expertise gaps. Those principles also highlight the need for gender-balanced boards (40% women, 40% men, 20% any gender) to stop groupthink and make better decisions. Cyber-attacks, data protection, and AI are explicitly listed as foreseeable risks that companies must now disclose.

You see similar moves in the G20/OECD Principles (revised 2023/2024). They stress specialised committees and thinking about stakeholder interests. They advocate for digital transformation strategies that are “digital by design,” embedding tech into every stage of policy. These changes mean directors must be “fluent with technology.”

They need to be able to challenge conventional thinking. That requirement for “cyber expertise” on boards has shot up by 86% since 2019, which really tells you how critical these skills are becoming for modern governance.

The Way Forward

Bringing in Digital Non-Executive Directors isn’t a luxury for just the tech companies anymore. It’s a fundamental requirement for every board, really. That digital knowledge gap creates friction, it creates underinvestment, and it creates systemic risk that can genuinely jeopardise a company’s future. By shaping the right questions, a Digital NED takes technology from being an operational cost to a true strategic asset.

To be effective stewards, boards typically have to focus on a few things:

• Substance matters more than status. Prioritise technical insight and strategic vision when recruiting, not just reputation.

• Break that vicious recruitment circle. Intentionally define your digital strategy first, and then use that to drive targeted hiring.

• Govern AI ethically. Define your AI posture. Put policies in place that prioritise transparency, fairness, and accountability.

• Manage the debt. Treat technical and digital debt as a strategic risk. Allocate resources for modernisation alongside innovation.

• Encourage challenge. Use the digital expertise to actively counter groupthink. Make sure every major strategic move is thoroughly tested.

As regulatory expectations on “material internal controls” and “cyber expertise” keep climbing, boards that don’t adapt will find themselves increasingly exposed. Exposed to being overtaken by competitors, exposed to legal liability.

The message is that the most effective boards are those where digital literacy isn’t some optional extra. It’s a core competency, just as essential as knowing your way around a set of accounts or understanding legal nuance. And that new standard of board accountability, well, it starts with the recruitment of Digital NEDs who have the guts and the competence to truly shape the future.