# Research taste lives in specifically non-obvious territory <div class="pills-container"><span class="pill">Last Updated: April 2026</span></div> Research taste is roughly the set of intuitions that guide good decisions throughout a project — which experiments to run, when an anomaly is signal versus noise, what the core communicable claim is inside a messy set of findings. What's characteristic about it is that it operates specifically in the domain of things that aren't immediately obvious. If a question were immediately obvious, you wouldn't need taste to answer it — you'd run the obvious experiment. Taste is deployed when the right move isn't clear. This is also what makes [[You can't grind toward research taste because feedback loops are too slow|grinding toward taste ineffective]]: you can't practice in non-obvious territory by repeating things in obvious territory. And it's why [[Developing research taste requires deliberately creating feedback|creating feedback deliberately]] is necessary — the non-obvious territory doesn't produce feedback automatically.