UX Glossary · H
Heuristic evaluation
Also: Expert review
Experts reviewing an interface against established usability principles to find problems without recruiting users.
Heuristic evaluation is an inspection method: evaluators examine an interface against a set of recognised usability principles and record where it violates them. Jakob Nielsen and Rolf Molich introduced it in 1990, and Nielsen refined the accompanying ten heuristics in 1994.
Its real advantage
Speed and cost. No recruitment, no scheduling, no incentives. An experienced evaluator can review a substantial product in a few days and produce a ranked list of problems, which is why it is the backbone of most UX audits.
Nielsen’s guidance was that a single evaluator finds only a minority of the problems present and that three to five evaluators working independently find substantially more, because different people notice different things.
Its real limitation
It finds violations of principles, not surprises. An expert can tell you a label is ambiguous; only a user can tell you they thought it meant something entirely different and abandoned the task.
It also produces false positives. Something that offends a heuristic may cause no trouble in practice, and an evaluator with no domain knowledge can flag a convention that specialists rely on.
In practice
The strongest sequence is evaluation first, testing second. The review removes the obvious problems cheaply, so the expensive sessions with real users are spent on the questions only users can answer.