A post about culture goes around every few months, and it came past me again last week: culture is not what you say, it is what you tolerate; what you permit, you promote. It is sharp, quotable, and built to be screenshotted, and every time it lands in my feed I nod along for a sentence and then find myself arguing with the whole shape of it. Not because I have leadership solved. Closer to the opposite.
It is not that the post is wrong; the observation under it is real. A top performer who gets a pass nobody else would get does erode something, the quiet belief that greatness on this team is even reachable. But my disagreement comes from a humbler place than the post does. That version of culture flatters the leader, casting you as the bouncer at the door, the one thing standing between the team and chaos; what I keep learning, mostly from carefully watching the teams that felt smooth and effortless in their delivery, is that it was less about what I did and more about what the individual contributors on the team did. The post is not wrong, then, just half the picture: all of its wisdom goes to what a leader should refuse to tolerate, and none to what a leader should grow. You cannot punish a team into being great. You can only punish it into being quiet.
Every rule is a confession
Here is something I believe and can only half-defend: whenever you have to write a rule, it is because you already failed. The rule is a patch stuck over the exact spot where the wrong thing was easier than the right thing. "Assume good intent." "There are no dumb questions." "Give credit where it is due." Every one of those is an admission that, left to the path of least resistance, people did the other thing, and rather than move the path you printed a sign and checked a box.
The good version of a rule is almost never a poster; it is a change to the terrain. You do not put "give credit" on the wall and hope; you make the person who did the work the one who stands up and presents it, so the credit has nowhere else to go. My own version of that is a credo I have led by for years: we have succeeded, I have failed. When the work lands, it was the team; when it goes sideways, and eventually it always does, that one is on me. It is not modesty; it is terrain. If credit reliably flows down and blame reliably stops at my desk, people will take the swings that real growth requires, because I have made the downside safe and left the upside theirs.
That is the move. Make the correct thing the default, the easy thing, the thing that happens when someone is tired and just wants to go home, and you do not have to tolerate or not-tolerate anything, because the behaviour you wanted is simply what falls out of the path of least resistance. Enforcement is what is left over when you could not be bothered to fix the terrain.
AMP, and the trust it sits on
So if not enforcement, then what? The honest answer did not come from a management book. It came from a parenting book about toddlers, of all places (Joanna Faber and Julie King's How to Talk So Little Kids Will Listen), which is really one long argument for listening to a small person instead of managing them. The frame I took from it has never left me: autonomy, mastery, purpose. AMP. Autonomy is handing someone the task and genuinely trusting them to build something good; that trust is not a courtesy, it is where their confidence starts. Mastery is giving people the room to learn and grow: space to explore an idea, to float a proposal and not have it shot down out of hand. And purpose is the belief that the work actually matters, that a thing gets built because there is a hypothesis and a real bet on value behind it, not because someone needed to be kept busy. Feed those three and people do not need to be policed, because they are already pulling in the direction you wanted.
But AMP does not run on good vibes and a values deck; hope is not a strategy. It runs on trust, and specifically in a room where people know they can say the thing: the half-formed idea, the disagreement, the quiet "I think this is actually wrong," without getting bitten for it. A genuinely free space. That is the soil; autonomy, mastery, and purpose are what you plant in it. Skip the soil and none of the rest takes, because a person who does not feel safe saying the true thing will just say the safe thing, and now you are back to a quiet team, which the enforcement crowd mistook for a healthy one.
A quiet team is not a healthy team. It is a frightened one that has learned to say the safe thing.
Confidence is a flappy bird
There is a variable underneath all of this that never makes it onto the values wall, and it is confidence. Not arrogance, and not a fixed trait you either have or you do not. Confidence is a level, and the level never holds still. It behaves, and I am only half joking, like Flappy Bird. Gravity is constant and it is always pulling down. Left alone, confidence sinks: the deploy that failed, the review that came back brutal, the feature nobody used, the imposter voice on an ordinary Tuesday. It falls, and it keeps falling, until something gives it a bump: a shipped thing that worked, a hard bug actually resolved, praise from someone whose opinion you value. Then it rises, clears the pipe, and gravity takes back over and it starts sinking toward the next one.
Which means praise is not a nicety you hand out when you happen to remember. It is the stimulus. It is the tap that keeps the bird in the air long enough to clear the next gap, and a team full of people whose confidence is quietly on the floor is a team that has stopped taking the swings that grow anyone. The catch is that the bump has to be real and specific to land; everyone can feel the difference between "good job team" and "the way you went and killed the twenty-minute build the whole team had just learned to live with, no ticket, nobody asking; that is the part I never had to assign." One is noise. The other is fuel, because it tells the person exactly which thing they did was good, so they, and everyone watching, can do it on purpose next time. That is the same signal I said I hunt for when I interview; the difference is that in the interview I am measuring it, and on the team I am trying to grow it.
Only works on the willing
Do this across a whole team (AMP, the trust under it, the confidence bumps landed at the right moments) and the culture drifts upward more or less on its own. It is slow. It moves on the timescale of seasons, not sprints, the same as judgement does, and for the same reason: you are growing something, not installing it.
But here is the catch, and it is the one place the tolerance post and I actually shake hands, just reaching from opposite ends. All of it (AMP, trust, the bumps) only works on people who wanted to grow in the first place. You cannot AMP someone who is phoning it in. Autonomy is wasted on them, mastery is not something they were chasing, and purpose bounces off. And they do not just fail to rise; they sap the room. Energy on a team is finite, and one person coasting on everyone else's effort is a slow leak that drags every bird in the flock down toward the pipe. That is the rot the original post was actually pointing at, and it was right that it spreads; it was just wrong that the answer is a bouncer.
The answer is upstream of all of it. Every leadership philosophy, mine very much included, eventually comes down to one unglamorous decision: who you let in the door. Hire people who want to build something and be part of something, and AMP has something to work on. Hire people who want a paycheque and a quiet life, and no amount of culture-tending saves you; you will spend your leadership standing at the edge of the room refusing to tolerate things, which is to say you will have proven the post right by hiring in a way that made it true. So, fine: what you permit, you promote. But culture is not what you forbid; it is what you make easy. What you celebrate, you cultivate. And what you hire is what you get.
