When you say "listen to people's lived experiences!" I completely agree. And I'll even one-up you: we should undertake to collect a wide, representative sample of lived experiences, carefully controlling for potential sources of bias, and report those experiences robustly, with charts.