UX Design Fundamentals · 18 min · 130 XP
Heuristic evaluation
Inspect an interface against Nielsen's ten usability heuristics to catch common problems cheaply, and know what this can't tell you.
A heuristic evaluation is an expert review against a short list of principles. It costs an afternoon, needs no participants, and reliably finds the common problems, so a usability test can spend its five precious sessions on the problems only real users reveal. The list most teams use is Jakob Nielsen's ten usability heuristics, published in the 1990s and still in use because they describe how people behave, not how screens look.
1 Visibility of system status 2 Match between the system and the real world 3 User control and freedom 4 Consistency and standards 5 Error prevention 6 Recognition rather than recall 7 Flexibility and efficiency of use 8 Aesthetic and minimalist design 9 Help users recognize, diagnose, and recover from errors 10 Help and documentation
A few are easy to mix up. Visibility of system status means people always know what's happening: a save that shows no feedback, or an upload with no progress, breaks it. User control and freedom is the way out: undo, cancel, back, an exit from a flow entered by mistake. Error prevention stops the mistake happening at all, such as greying out a date that can't be booked. Recognition rather than recall means showing options instead of making people remember them: a list of recent files rather than asking them to type a path.
How to run one: have three to five people review independently, each going through the key tasks twice (once to get the feel, once to inspect). For each problem, note where it is, which heuristic it breaks and how severe it is. Only then merge the lists. Independent reviews matter: each evaluator finds a different subset, and discussing too early makes everyone find the same things.
Experts aren't users. A heuristic evaluation finds violations of known principles. It can't tell you whether people want the feature, understand your vocabulary, or would ever click that tab. Treat it as a filter before testing with real users, never as a replacement for it.
Do it this week: review one flow of a product you use against the ten heuristics. For each problem, note the screen, the heuristic and a severity from 0 to 4. Then ask a colleague to do the same alone, and compare lists.
Loading your workspace…