Textbook of Usability

Heuristic evaluation, developed by Jakob Nielsen and Rolf Molich in 1990, is a systematic inspection method in which trained evaluators examine an interface and judge its compliance with established usability principles (typically Nielsen's 10 heuristics). Evaluators record every problem they find, noting which heuristic it violates and rating its severity.

The standard procedure:

1. Brief evaluators on the user population, tasks, and context
2. Individual evaluation — each evaluator independently examines the interface at least twice
3. Document problems — location, heuristic violated, likely impact
4. Consolidate findings across evaluators
5. Rate severity — combining impact and frequency

A critical principle: multiple evaluators are essential. Individual evaluators find different subsets of problems. Nielsen recommends 3–5 evaluators, who together typically identify about 75% of usability problems. A single evaluator catches far fewer.

Strengths: fast (1–2 hours per evaluator), inexpensive (no participant recruitment), can be applied to wireframes or prototypes, produces actionable findings.

Limitations: depends on evaluator expertise; reports false positives (problems real users wouldn't encounter); misses problems arising from user mental models rather than heuristic violations; does not produce performance metrics or satisfaction data.

Heuristic evaluation is most effective when combined with user testing — the two methods identify overlapping but distinct sets of problems.