Heuristic evaluation

What a heuristic evaluation checks

Most teams use Jakob Nielsen's 10 usability heuristics. The first five are visibility of system status, match with the real world, user control and freedom, consistency and standards, and error prevention. The rest are recognition rather than recall, flexibility and efficiency, minimalist design, helping users recover from errors, and help and documentation.

They're broad rules of thumb, so they apply to almost anything: a SaaS dashboard, a checkout, a voice interface, even a rough prototype. A file upload with no progress indicator breaks the first one. A form that wipes your answers after a validation error breaks several.

The 10 usability heuristics, in plain words

Jakob Nielsen published the ten heuristics in 1994, and Nielsen Norman Group still maintains them. The names are his. The one-line explanations are ours.

• 1. Visibility of system status. The product always shows what is happening: saving, loading, sent, failed.
• 2. Match between the system and the real world. It uses the words and the order your users use, not internal terms.
• 3. User control and freedom. People can undo, cancel and go back without losing their work.
• 4. Consistency and standards. The same word and the same control mean the same thing everywhere, and follow what people know from other products.
• 5. Error prevention. The design stops mistakes before they happen, for example by confirming a deletion.
• 6. Recognition rather than recall. Options are visible, so people do not have to remember them from an earlier screen.
• 7. Flexibility and efficiency of use. Shortcuts and defaults speed up experienced users without confusing new ones.
• 8. Aesthetic and minimalist design. Each screen shows what is needed for the task and leaves out the rest.
• 9. Help users recognise, diagnose and recover from errors. Error messages say in plain language what went wrong and what to do next.
• 10. Help and documentation. Help is easy to find, tied to the task and short, for the cases where it is needed.

The full wording and examples are in Nielsen Norman Group's article, 10 Usability Heuristics for User Interface Design.

How to run a heuristic evaluation, step by step

Pick a narrow scope: one task, one user group, one device. Then have three to five people review it independently, which is what Nielsen Norman Group recommends, because any single evaluator misses things. Timebox each review to an hour or two and keep notes private until everyone has finished.

Each evaluator records the problem, where it occurs and which heuristic it breaks. Then the group merges notes, clusters duplicates and rates severity. The final list should say what to fix first and which questions need a usability test to settle.

Why severity ratings matter

Ten broad rules applied to every screen produce a long list. Much of it will be small: a label that is slightly unclear, an icon without a tooltip. A few items will be the reason people give up.

A flat list hides that difference, and a team that gets forty findings with equal weight tends to fix the easy ones. So each finding needs a severity rating. Nielsen's scale runs from 0, not a problem, to 4, a usability catastrophe, and weighs how often the problem occurs, how hard it is to get past and whether it keeps happening.

Rate after the evaluators merge their notes, not before, so the rating reflects the group and not one person's irritation.

Heuristic evaluation vs usability testing

A heuristic evaluation needs no participants, so it's cheaper and quicker to run, and it works on designs that aren't built yet. Usability testing watches real people attempt real tasks, which is the only way to see how your particular users think.

Nielsen Norman Group is clear that heuristic evaluation can't replace user research. Use it to clear out obvious problems and to decide what a later usability test should focus on.

The two methods have their own entry here: usability testing.

Common heuristic evaluation mistakes

Relying on one evaluator is the big one. You get one person's blind spots presented as a full review. Another is writing findings like "violates heuristic 4" with no screen, no example and no suggested fix.

The subtler mistake is treating every breach as a defect. Heuristics are guidelines, not laws, and some breaches are deliberate trade-offs. The job is to judge whether the breach actually hurts users in this context.

A last one is scope creep. An evaluator who starts on the checkout and wanders into account settings produces a shallow review of both. Agree the task before anyone opens the product, and note anything outside it for another day.

How Gr8r uses heuristic evaluation

Heuristics are one of the lenses in our UX audits, alongside established interaction patterns and whatever data you can share. We use them to explain why something is a problem, not as a checklist to score.

Every finding is tied to a specific screen and rated by severity. Each one comes with a recommendation your team can act on.

A UX audit is the wider piece of work, and our UX audit service is free for the first 5 companies that ask: request the free UX audit. For the screen by screen method, read how to do a website UX audit. A heuristic evaluation is not an accessibility review. For that, see accessibility audit.