The most dangerous assumptions are the ones we no longer realise we're making.

Somewhere in your business there is a rule nobody can explain.

A spending threshold. A sign-off step. A report that gets produced every Monday and read by nobody. A rule about which customers you will not take on. Ask why it exists and you will get one of three answers: a shrug, the name of someone who left years ago, or the sentence that should worry you most: "that's just how we do things here."

None of these things arrived by accident. Every one of them was, at some point, a good decision. Someone looked at a real situation, weighed it, and made a judgement. The judgement worked. Because it worked, it was repeated. Because it was repeated, it was formalised. And once it was formalised, it stopped being a judgement at all.

That is the sequence, and it runs quietly in every organisation I have worked in or led:

Then reality changes. And nobody notices.

Assumptions are not stupidity. They are compression.

It is tempting to read that sequence as decay, and to conclude that organisations should hold fewer assumptions. They should not. An institution that re-litigated every decision from first principles would never do anything. Assumptions are how a group of people act at speed without convening a meeting about the meaning of everything.

Compression is a feature. It is what lets a business of two hundred people behave with any coherence at all.

The problem is what gets discarded in the compression. A judgement always has two parts: the conclusion, and the conditions it was made under. We were right to price this way, because that competitor existed, because that channel was cheap, because customers bought in that pattern. The conclusion is portable. The conditions are not.

And it is always the conditions that get dropped first, because they are the boring part. What survives into the process document is the answer. What is lost is the world the answer was an answer to.

So the rule outlives the reason. Nothing announces this. There is no moment where a policy stops being right. It simply drifts out of correspondence with reality while continuing to look, from the inside, exactly as correct as it did on the day it was made.

Rigidity is a perception problem, not a will problem

We usually explain organisational rigidity as reluctance. People resist change. Leaders defend their territory. Culture is slow. There is truth in all of it, but it flatters us, because it locates the failure in motivation - and motivation is something we know how to attack. Communicate the vision. Build the burning platform. Bring people with you.

That is a different diagnosis, with different consequences. You cannot renegotiate an assumption you cannot see. If the organisation genuinely believes the operating model reflects how the world works, rather than how the world worked in 2019, then every conversation about changing it will feel like a conversation about breaking something that isn't broken. The resistance is not obstruction. It is the entirely rational defence of a picture that no longer matches the terrain.

This is also why transformation programmes so often produce motion without adaptation. They act on the visible: structures, reporting lines, systems, titles. The assumptions doing the actual load-bearing are invisible, unwritten and unowned, which is precisely why they survive the restructure intact, and why the new organisation starts behaving like the old one within about eight months.

Where to look

There is no framework for this, and I would be suspicious of one. Noticing is not a process you install; it is a habit you protect. But there are places where the drift is unusually visible, and they cost nothing to inspect.

Start with the language. "We've always", "we don't", "that's just how": these are not explanations. They are the sound an assumption makes when you touch it. Every one of them marks a decision that has stopped being reviewable. Not one of them tells you when the decision was made, or by whom, or on what evidence.

Then date your rules. Almost nothing in an organisation carries a timestamp. Take any recurring practice and ask two questions: when did we start doing this, and what was going on when we did? The answer is often startling. A pricing floor set during a supply shortage. An approval step added after one incident nobody now remembers. A meeting rhythm designed for a company a third of the current size. None of it wrong when it was made. All of it unexamined since.

Then run the inversion. Rather than asking "is this still working?", a question that invites the answer yes, because the thing is producing numbers and nobody has complained, ask what would have to be true for this to still be the right call. Say the conditions out loud. Then check whether they hold. This is a much harder question to answer comfortably, which is the point.

And finally, treat new people as an audit you are being given for free. In their first ninety days, they can see the absurdities plainly, because they have not yet been taught which questions are naïve. Most organisations spend that window carefully explaining the reasons for everything until the newcomer stops asking. That is not induction. That is the transmission of assumptions, and it is the fastest mechanism of institutional inheritance we have.

Why this is the foundation, not a footnote

I have spent a lot of time on how organisations become more adaptable, on how work gets shaped, how decisions move, how businesses stay light enough to change direction without a reorganisation. Underneath all of it sits something more basic.

No institution can consistently make good judgements beyond the quality of its own perception. If your picture of reality is out of date, speed does not help you. Neither does capability, or talent, or better data. You will simply execute the wrong thing more efficiently. This is why adaptability begins with noticing, and not with change, because change acted on a stale picture is just expensive motion.

The organisations that stay adaptable are not the ones that reorganise most often. They are the ones that keep a smaller proportion of their beliefs in the "settled" pile. They can still tell you why they do what they do, and the answer is about now rather than about 2018.

So the question worth carrying is not "what should we change?" That question comes second and gets far too much attention.

Most of us cannot answer that immediately. That difficulty is not a failure. It is the finding.