See what's really there.

Every organisation runs two structures at once. The first is the one on paper, reporting lines, job titles, governance diagrams. The second is the one people actually use: who everyone goes to when something goes wrong, where information gets stuck, who quietly holds everything together, whose opinion a decision genuinely waits for.

The gap between the two is where most strategic problems live. Influence rarely sits where the chart says it does. The person three levels down with twenty years of memory may carry more decision weight than the person with the senior title. None of this is visible from the leadership off-site.

Before you redesign anything, map what is real. Ask who people actually go to, what actually gets decided in corridors, and which meetings exist purely to ratify what was agreed elsewhere. Strategy built on the paper organisation will keep being defeated by the real one.

Chronic busyness reads like a resourcing problem, so organisations reach for the resourcing solution: more people, more budget, more hours. But stretch is a symptom with many possible causes, and headcount only fixes one of them.

Sometimes the issue is priorities, too many things being treated as important, which is the same as nothing being important. Sometimes it's decision rights: work stalls because nobody is quite sure who can say yes. Sometimes it's duplicated effort, or processes that generate work about work. Adding capacity to any of these just makes the confusion more expensive.

The diagnostic question is not "do we have enough people?" but "what is consuming the people we have?" Until you can answer that honestly, new resource disappears into the same fog.

Every organisation has two strategies. The declared one lives in the plan, the away-day output, the investor deck. The real one lives in the budget, the diary and the incentive structure. When the two disagree, the real one wins, every single time.

This is why reading resource allocation is such a reliable diagnostic. If the plan says innovation but the budget says maintenance, the strategy is maintenance. If the plan says customers but leadership attention goes to internal politics, the strategy is politics. Money and time don't lie about priorities; documents do it constantly.

The uncomfortable exercise: hand someone your strategy document and your management accounts, and ask them to tell you what your strategy is. If the two answers match, you're rare. If they don't, you have a choice about which one to change.

Healthy organisations argue well. Ideas get stress-tested before the world stress-tests them expensively. So when a leadership team reaches quick, warm consensus around the leader's view, the likeliest explanations are not "the idea is excellent" but "people have learned the cost of disagreeing" or "people have stopped believing disagreement changes anything."

The leader is usually the last to know. Every time a challenge gets brushed past, every time the most senior voice speaks first, the room recalibrates what is sayable. Eventually the leader is operating on filtered reality, the most dangerous information environment there is, because it feels like clarity.

The corrective is deliberate friction. Speak last. Reward the person who finds the hole. And treat suspiciously easy agreement as a finding to investigate, not a win to enjoy.

Longevity is not evidence. A process that has survived for a decade may have earned its place, or it may simply have never been made to justify itself. "We've always done it this way" describes history, not value. It answers "how long?" while pretending to answer "why?"

Every inherited process deserves an occasional trial. What problem was this built to solve, and does that problem still exist? What would happen if we stopped? And the sharper question: who benefits from this staying exactly as complicated as it is? Complexity is rarely neutral, it usually protects someone's territory, budget or relevance.

Sunset clauses are underrated in organisational life. Anything that can't periodically earn the right to exist should lose it.

The things organisations are proudest of are the hardest to examine. The flagship programme, the founding product, the famous culture, these become identity, and identity resists interrogation. Questioning them feels like disloyalty, so nobody does, and the asset quietly becomes an anchor.

This is how success produces rigidity. The capability that made you distinctive in 2015 may be the thing stopping you moving in 2026, not because it's bad, but because everything else has been arranged around it. Resources, structures, self-image: all optimised for a strength whose moment may have passed.

The question isn't "is this still good?" It's "what is this costing us that we can't see?" Pride is not a strategy review.

When a problem survives multiple well-funded attempts to fix it, ask a different question: what would a real solution require someone to admit? Problems that trace back to a senior person's decision, the restructure that didn't work, the system that shouldn't have been bought, the hire that isn't working, develop a strange immunity. They get managed, worked around and studied, but never solved.

This isn't conspiracy; it's self-preservation. Solving the problem means naming its origin, and naming its origin has a cost. So organisations buy workarounds instead, new layers, new roles, new initiatives, each one cheaper than honesty and more expensive than the fix.

The diagnostic: who would be embarrassed if this were actually resolved? That's usually where the real work is.

"People don't like change" is one of the most convenient myths in organisational life, convenient because it blames the audience rather than the proposal. The same people labelled resistant adopt new phones, new habits and new employers without ceremony. What they resist is change they haven't understood, haven't influenced, don't trust, or can't see making their work better.

Each of those is a different failure with a different fix. Not understood is a communication problem. Not influenced is a respect problem. Not trusted is a history problem, and you cannot communicate your way out of a trust problem, because the channel itself is what's broken. Not seeing the benefit is a design problem.

Before diagnosing resistance, ask which of these your change has actually earned. The answer is usually in the design of the change, not the character of the people.

There is a moment, after the town hall and the slide deck, when leadership teams feel the change has happened. It hasn't. What has happened is an announcement. The change happens, or doesn't, in the thousand small decisions people make afterwards, most of them invisible to the people who made the announcement.

This gap explains the graveyard of initiatives that were launched and never landed. Announcing is easy and visible; embedding is slow and unglamorous. It means answering the same question for the fifth time, changing the systems that still reward the old behaviour, and noticing that the incentive structure quietly contradicts the new direction.

If your change plan ends at communication, you don't have a change plan. You have a press release.

Founders succeed through personal involvement, every decision, every relationship, every standard passing through one pair of hands. That total ownership is the engine of early success. It is also, later, the constraint. An organisation that only works when the founder touches it can only ever be as big as one person's attention.

The painful part is that the thing being given away isn't a weakness; it's the original strength. Letting others own decisions the founder would make differently, watching standards be interpreted rather than enforced, being left out of rooms, this feels like decline from the inside. It's actually scale.

The founder's final product is an organisation that no longer needs them to function. Building that is the hardest act of creation there is, because it means undoing, carefully and deliberately, what made everything else possible.

Most organisations are far better at starting things than stopping them. Initiatives accumulate; almost none are ever formally buried. The result is a sediment of zombie projects, too dead to succeed, too politically protected to kill. Stopping requires someone to say, out loud, "that didn't work", and in many cultures that sentence is career-limiting.

But that sentence is the price of learning. Every honest post-mortem teaches more than a dozen success celebrations, because failure contains information and success mostly contains confirmation. An organisation where nothing is ever allowed to have failed is an organisation that learns nothing, it just accumulates.

Knowing when to stop is not defeat; it's resource allocation. The courage to end something well is what frees the energy to start the right thing next.