Changing Company Culture: Why It Usually Fails and What Works

Culture change has a bad reputation, and it is deserved. The usual programme looks like this: a leadership session, five new values, a kick-off, posters, an intranet post. Six months later nothing has changed except the wording on the wall.

That is not because the values were wrong. It is because culture does not live at the level of values.

Why it usually fails

Culture is the pattern of shared assumptions about what works here — and assumptions do not change because someone on a stage says something different. They change because people see new evidence.

Three predictable reasons programmes stall:

The reward system did not move. An organisation claiming to shift toward collaboration while rewarding only individual targets has taught its people that the new values are decorative. People believe the bonus structure, not the poster.

No visible proof was supplied. Culture change becomes believable the moment someone demonstrates the new behaviour and is visibly better off for it — or demonstrates the old behaviour and is visibly held to account, including the people who could previously afford not to be.

It is one event rather than a frequency. One kick-off changes nothing. Twenty small consistent signals across a quarter do.

What does work

1. Pick one behaviour, not five values

'More open communication' is not executable. 'Problems get raised in the week they appear, not the month after' is. One concrete, observable behaviour at a time. If you cannot see whether it is happening, you cannot change it.

2. Measure the baseline before you start

Without a baseline, every discussion about progress is a battle of opinions. It does not need to be a large survey — one short pulse question per team about that one behaviour is enough, provided you repeat it.

What you want is the line per team over time, not a single number for the whole organisation. The company-wide average is almost always the least useful figure: it hides that three teams are fine and one is in trouble.

3. Change what visibly gets rewarded

This is the hinge. People infer the real rules from who gets thanked, who gets promoted, and what goes unaddressed.

The cheapest lever is public recognition. When one colleague is visibly thanked for raising a problem early, the rest of the team has learned within a week that doing so is safe here. No memo achieves that.

4. Have leaders go first, including on being wrong

Psychological safety does not come from a manager saying mistakes are allowed. It comes from the manager naming their own mistake first and not coming out of it looking better. Until that happens, it stays an offer nobody dares accept.

5. Close the loop, every time

The fastest way to kill culture change is to measure without responding. If people answer and visibly nothing follows, they learn the survey is theatre — and next round they answer less honestly, or not at all.

So after every measurement say three things: what we heard, what we are doing about it, and what we are deliberately not doing and why. The last one matters more than it looks.

In remote teams this is harder, and more important

In an office you can read from the room whether a change is landing. At a distance you cannot. What the office gave you free — noticing who goes quiet, hearing how something lands — has to be explicitly replaced remotely by frequent, small signals.

That is also the only honest answer to 'how long does it take'. Not one quarter, and not 'until the kick-off'. It takes until the new behaviour has been rewarded often enough that people start expecting it. In practice that is months of repetition — most of it small, boring consistency rather than grand gestures.