“Once people begin to act they generate tangible outcomes in some context, and this helps them discover what is occurring, what needs to be explained, and what should be done next. Managers keep forgetting it is what they do, not what they plan that explains their success. They keep giving credit to the wrong thing-namely, the plan-and having made this error, they then spend more time planning and less time acting. They are astonished when more planning improves nothing: Karl Weick, Sensemaking in Organisations, 1995”
Essentially, doing stuff teaches you stuff. The more you do, the more you learn. But only if you have theory. If you have theory, then you can learn. You explicitly test theory against reality and based on the results you adjust your theory.
If we do X then Y will result. IF I rub the potato with salt and olive oil before putting it in the oven THEN the skin will be extra crispy. These are experiments. Experimenting in the work is the way to improve.You don’t know when you start what to do, if you did you would have done it already, but you haven’t, you’re just starting so you don’t know, you just have good ideas based on knowledge. The purpose of an experiment is to get more knowledge, to test your theory. Obviously sometimes you wont have the right theory, or an accurate theory anyway, and something else happens, you don’t get what you expected to get. But that’s OK, it didn’t fail. Experiments only fail if you don’t get knowledge, if you haven’t designed the experiment so that you can learn from it. If you don’t get knowledge, you didn’t design it right. But it doesn’t fail if the hoped for improvements aren’t realised. You’re one step closer, having ruled out one of the ways that don’t work.
Seth Godin says similar today in his post “The alternative to failure”
To the critic who hasn’t shipped, who hasn’t created his art, anything less than better-than-what-I -have-now appears to be a waste. To this critic, progress should only occur in leaps, in which a fully functioning, perfected new device/book/project/process/system appears and instantly and perfectly replaces the current model.
Of course the unstated assumptions that you not only can but should get it right first time is ludicrous. It is ludicrous to expect perfect new improved systems to spring fully formed into reality having missed out that annoying stage of not working, or being half formed or half thought through, missing vital elements not yet imagined. Tim Harford’s new book, Adapt, explores the failure of the grand 5-year plan, the Stalinist attempts to shape future reality through implementing written documents. He reveals the importance of adaptation, rapid prototyping and testing to build up better products, services and ideas. Just like with evolution, there are far more losers than winners. Like when generating ideas, the best way to produce a great idea is to produce a thousand ideas, 999 of which are rubbish, but you still have to produce the 999. You cannot pick and choose in advance.