I don't inherently hate A/B testing. I've just seen so many clients miss the forest for the trees because of it. We had a client who tested every design variation of their UI down to the color and label of buttons. They were able to tell us that blue buttons converted better than green ones. Why? No idea. Why? Because we don't talk to customers. Why? It doesn't even take five tries until you get to the root of the problem in companies that push all of the design responsibility to the end of the line saying "we'll just test it in production." I applaud their desire to get good data from real people, but the clapping slows to golfing pace when I hear them defer everything to A/B testing in production. That is the most expensive form of research in time, money and the effort to unwind and rewire what you just launched. The dirty reality is that people who often say "we'll just A/B test it" know that it'll never get unwound or reworked. So do yourselves a favor and instead of kicking the design responsibility can down the production, take a beat. Think. Decide. What should you be designing and why?
How do you separate the designs you can do from the designs you should do? Hint: the answer isn't to do them both and let an A/B test sort it out. That mentality sets your design practice back a few decades where teams tested every plausible design idea because it was undiscovered territory. Since then, design has evolved for most of us. We use design strategy to guide us, erect guard rails to focus our thinking, and to illuminate territories worth exploring. Design strategy is the answer. #designstrategy #mindfulchoices
