UX

Run critique against named principles, not taste

Every critique comment is anchored to a named law, heuristic, or guideline so feedback becomes reviewable rather than personal.

How to run it

  1. 01

    Open critique by restating the user goal and the decision the team actually needs to make.

  2. 02

    Require each comment to cite a principle - Hick’s Law, a Nielsen heuristic, a WCAG criterion, or a HAX guideline.

  3. 03

    Log unresolved disagreements as assumptions to test, not as opinions to settle in the room.

  4. 04

    Use Claude or ChatGPT to pre-generate a heuristic checklist for the specific screen type before the session.

  5. 05

    Close by naming the decision, the owner, and the evidence that would change it.

Anti-pattern this prevents

Critique that resolves on seniority. The loudest or most senior opinion wins, the rationale is never recorded, and the same argument recurs next sprint.

Supporting tools