Learning from mistakes, make mistakes, learn, it’s OK, mistakes are OK, we are human, if we did not make mistakes we would not innovate, in this company you can make mistakes, we learn, we learn, just make sure the mistake is proportionate and you take prudent risks. Incidentally, I wonder if you have heard of our brand new initiative on Excellence, you know, do it right the first time and all that.
That script is not unusual in an organization where management feels compelled to say ’learn from mistakes’ but (1) it’s terrified of your mistakes and (2) there are plenty of examples when people who made mistakes, even the ‘prudent ones’, did not have a particular good time. In fact, the ones at the top seem to be those who never made a mistake, not because they were particular smart but because they never put themselves in a position to make them.
The only decent way to shape a culture of openness and true ‘learning form mistakes’ is to (1) publicise them and (2) make risk/mistakes ‘symmetrical’ ( a la Nassim Taleb’s ‘Skin in the Game’)
- Publicise. Ten years ago, 2008, in Disruptive Ideas [1] , I suggested a ‘Hall of Fame of Mistakes’ in which mistakes are demystified and made public so that their emotional charge decreases when you can see all those fellow human beings also getting things wrong . It’s not good enough for the manager to know. Others must. Other must also publicise theirs. Don’t get too fundamentalist about it. I am not suggesting a semi-Maoist public disclosure ending in a rehabilitation camp. Do it humane and fun. It works.
- Make it ‘symmetrical’. If you post your mistakes, your manager must post his as well. We are not talking ether about cataclysmic mistakes. It may go as simple as that: This month I got wrong the assumption that X competitor was not ready. Opps! They were!. BTW, Peter, what is your mistake or miscalculation of the month? Sure you have one!
Then it all ends as ‘what we have learnt’.
Avoid platitudes of the type ‘we must be more customer centric’. That means zero. I mean, nothing.