Graham McNicoll

For years I chased my team for experiment write-ups after the fact. It never worked, and eventually I understood why. You ask someone to document a test at the exact moment they have the least reason to care. The results are in, that was the fun part, and now you want them to stop, switch to another system, and fill out a form for a future teammate they will never meet. So they don't- and it's like herding cats to get them to do it. The doc either never gets written or gets written badly and never read. The timing is backwards. The moment of real leverage is when someone wants to turn a test on. They are motivated, the hypothesis is fresh, and they will happily write down what they are trying to learn. Capture it there. The other half is removing the step. Your tool should already hold your hypothesis, your screen shots, your metrics, your results. Recording the learning should be one flow you are already in. Nobody wants a second place to keep notes- and those are usually hard to use. This is the reason we called the company GrowthBook, btw. Most tools in our space don't do this well - and a test you cannot learn from later is only half an experiment. We rebuilt a lot of this into 5.0. If your team has cracked getting learnings written down without anyone nagging, I would like to hear how. https://lnkd.in/gqvRBmym