User-Testing articles and design guides

Watching five people use your thing beats a month of internal debate. That's the spirit behind this tag: usability sessions, unmoderated tests, how to write tasks that don't lead the witness, and reading results without cherry-picking the ones you like. We get into recruiting, sample sizes, and the trap of testing your own bias back at yourself. Designers and PMs who want evidence instead of hallway consensus will find methods they can run cheaply and fast. Nielsen's five-user rule gets a fair hearing and a few caveats here, because it's more nuanced than the slide decks admit.