Heuristic evaluation is an expert review of an interface against a set of established usability principles — the “heuristics.” A few evaluators inspect the product, flag where it breaks a principle, and rate how serious each issue is. It’s fast, cheap and doesn’t need users in the room, which makes it the workhorse of most UX audits.
It won’t tell you what real users will do — only expert judgement about likely problems. But run well, it surfaces the majority of obvious usability issues in a day or two, so you fix the easy things before spending on user testing.
The usability heuristics
Most evaluations use Jakob Nielsen’s ten heuristics, or a close variant. In plain terms:
- Visibility of system status — always show what’s happening
- Match to the real world — speak the user’s language, not the system’s
- User control & freedom — easy undo, clear exits
- Consistency & standards — follow platform and internal conventions
- Error prevention — stop mistakes before they happen
- Recognition over recall — make options visible, don’t rely on memory
- Flexibility & efficiency — shortcuts for experts, simplicity for novices
- Aesthetic & minimalist design — no clutter competing with the essentials
- Help users with errors — plain-language messages and a way out
- Help & documentation — available when needed, easy to search
How to run a heuristic evaluation
Three to five evaluators is the sweet spot — one person misses too much, and returns diminish past five. Each reviews the product independently first, then findings are pooled so nobody anchors on someone else’s opinion.
- Agree the scope and the key user tasks to walk through
- Each evaluator inspects independently, noting issues against the heuristics
- Combine findings and remove duplicates
- Rate each issue’s severity, then prioritise
Rating severity
A finding without a severity rating is just an opinion. Score each on how much it hurts and how often it bites, so the team fixes what matters first rather than the loudest complaint.
- 0 — not a usability problem
- 1 — cosmetic; fix if time allows
- 2 — minor; low priority
- 3 — major; important to fix
- 4 — catastrophe; fix before release
Heuristic evaluation vs usability testing
They answer different questions and work best together. Heuristic evaluation is expert opinion — fast and cheap, great for catching obvious issues early. Usability testing is real behaviour — slower and costlier, but the only way to know what actual users do. Smart teams run a heuristic evaluation first to clear the easy problems, then test with users to find the ones experts can’t predict. Both sit inside the wider UX audit checklist.
Used this way, heuristic evaluation is the highest-leverage first step in improving an experience: a day or two of expert review that turns a vague sense of “this feels clunky” into a prioritised, fixable list.
