Over peanuts and beer.

Why Most Culture Change Fails

I went for a drink with a friend a while back – dry roasted peanuts, a couple of decent pints of Japanese beer – and spent most of it listening to him describe a company that was, in his words, rotten to the core. Not struggling. Not, could be better. Rotten.

The whole organisation was sapping the life out of him and, as far as he could tell, out of everyone in it. He talked the way people tend to when they've stopped expecting anything to change: flatly, with a muted tone and a lack of human spirit. I bought him another beer and tried to lift his spirits with a packet of posh crisps. I left feeling I'd been let in on a slow tragedy that was likely to be in the news headlines soon.

A month later we had another beer, and I was talking to a different man. Same face, same job, an entirely different level of energy running through him. The muted, sapped tone was gone, and in its place was enthusiasm, which I hadn't heard from him in years.

What had happened, I asked – expecting a new role or a lottery win.

A new leader had arrived. And the first thing this leader did – the thing my friend kept repeating – was cancel things. The change programme. The transformation programme. The whole plethora of impressively named initiatives that had run for years and improved nothing.

In their place he did something astonishingly simple: he named, clearly and accurately, the handful of behaviours he expected to see in the company. By all accounts he made it unmistakably clear to his executive team that they weren't optional behaviours, and then behaved that way himself, visibly, from the top down, in every single moment of work. My friend reckoned it took only two weeks for the entire climate of the organisation to change. The path was now clear. The energy, hope and spirit started to return. The climb back from decline had begun – and you could feel it before anyone had rewritten a single value on a single wall.

I think about that fortnight a lot, because it absolutely demolishes what most organisations believe about culture. They believe culture is something you announce and change through a programme: new values on the wall, new behaviours unveiled at the town hall, a project with a swish sounding name and a sponsor and a launch date. Research suggests that around seven in ten of those efforts change nothing and fail, because they've misunderstood what culture even is.

Culture is not what an organisation says it values. It's absolutely not what is written on the walls. It's not what they tell everyone. It's what people actually do – every day, under pressure, when the meeting's overrunning and someone has to make a decision, when nobody important is watching, or when the customer phones up to complain.

It's how people speak to each other when it's hard going. How decisions get made when time is short and the pressure is mounting. How mistakes get treated – as opportunities to get better, or as evidence against somebody. It's how people are spoken to, treated and rewarded everyday. Everything else – the posters, the purpose statements, the engagement survey – is not the culture. People plus their group habits and behaviours equal the culture.

Which is exactly why a place like my friend's can turn in a fortnight when the right person arrives, sets a new bar and holds a high standard. It's also the part that unsettles many people: you don't get the behaviours you expect. You get the behaviours you tolerate.

People are constant readers of power and influence. Whatever those with influence actually do becomes the standard definition of what's normal around here, and no amount of values-communication survives contact with someone senior doing the opposite.

My friend's new boss knew this inherently. It didn't need spelling out. He didn't demand that people behave differently, he said what behaviours he would like, and behaved in that way, consistently, where they could see him, over and over, until it was simply how things were done. And the other half of it was what he refused to let slide, especially in his direct leadership team – a deviation from what was expected wasn't accepted without feedback and correction and kindness. A tolerated behaviour becomes the standard – not because anyone decided it should be, but because silence and no correction reads as permission. Naming a lapse clearly and early with kindness and support, while it's still a lapse and not yet a pattern, turns out to be a kind thing to do for everyone, and the more effective one.

None of which means culture is just a matter of good people and bad people. Far from it. Behaviour is mostly produced by the system people are standing in, and a great deal of what looks like a character flaw is a perfectly human response to a broken system. Those muddled expectations, confusing direction, incentives that reward the opposite of what's expected, a boss who says one thing and does another. Change the behaviour without improving the system that generates it, and you get a brief improvement before the old habits creep back in. Which is why the deepest culture work isn't really a people problem at all. It's a design problem, and it starts with actually understanding the system you're inside. Understand the system and you can shift that at the same time as holding the high bar of behaviours.

People also move more readily when there's something worth moving toward, which the cancel-everything leader supplied the most of. Once the noise of the unfit programmes and initiatives was gone, a clear picture of what the place could become finally had space to be crafted. My friend told me that what carried the message to everyone wasn't a slide deck of facts and figures, it was the new leader telling remarkably funny, poignant and interesting stories about the culture he is aiming for and why it's good to be part of it. A good story does more influential work than a hundred A1 printed value statements on the wall, because it gives people a sense they belong in the story – a role to play in the narrative to be written – a reason to connect to the work, which is a powerful way to emotionally connect people to the task at hand.

The interesting thing about this story is the fantasy of the timeline that most people hold when it comes to change. Culture doesn't shift on a project plan to the milestones and deadlines given to it; it shifts through accumulation from trusted people – a thousand small signals about what gets rewarded, what gets challenged, and what gets ignored – until one day the norms have moved. You can't launch that. You can only set the signals, consistently, and let them add up. Which is, said with optimism and hope, not a transformation at all. It's cultivation: deliberate, tended rather than announced. And as this CEO proved, that doesn't always need to take a long time – at least to start planting those initial signals.

And you don't have to be the new CEO to start. Whoever you are in a group, you're contributing to what it tolerates – what you laugh along with, what you let pass, what you model when the pressure's on. My friend got lucky; someone with real power arrived and changed the climate in a fortnight. Most of the time it's slower than that, and gentler, and more spread out. But the mechanism is identical at every level, and it was never the programme. It was the behaviour, repeated, demonstrated, nudged and communicated, until it became the group habit.

That's the difference between culture change that's scheduled and announced, and culture change that's real. One is to be heard. The other is lived. And over peanuts and a decent beer, you can usually hear which one someone's actually in.