Product·
Identity resolution
How getuserfeedback.com connects the dots when the same person shows up with different identifiers — across devices, sessions, and the login boundary.
Research, ideas, and posts on user feedback and the work of asking well.
People judge how an interface looks in a fraction of a second. An embedded survey that doesn't look like the app around it gets judged as a stranger, so its look should come from the host by default.
Pick one primary color and our theming engine derives the rest of the survey's colors, dark variant included. The on-brand result is the default; the illegible one is hard to reach.
A dark mode is a sign a team cares how its product feels. A survey running in that product matches it automatically, for the same reason the dark mode exists.
Most churn is invisible. The users who leave quietly are the ones you most needed to talk to.
Every named product-market fit pitfall is a different label for the same mistake — building for someone who was not in front of you. The fix is the same too.
The famous engagement thresholds — Slack's two thousand messages, Facebook's ten friends in seven days — are symptoms of fit you already had, not levers you can pull. Using them as goals inverts cause and effect.
Every field you add to signup is a chance for someone to close the tab. Most products keep adding them anyway.
The 40% rule tells you when a group of users has product-market fit. It doesn't tell you how to get there — that's a different kind of knowing.
Every ask in your product can now carry your name, your photo, and your title. The unit cost of a personal ask drops to zero clicks.
A handwritten Post-it note doubled survey response rates in one famous experiment. Online, the photograph and the first name do the same thing.
Building a survey used to mean filling out a form and imagining the result. Now you edit the survey itself — what you see is what your respondent gets.
For every user who tells you something is wrong, twenty-five say nothing. Some of them are already leaving.