Jakob Nielsen's five dimensions, from Usability Engineering (1993): Learnability — how easy it is to accomplish basic tasks the first time. Efficiency — how quickly experienced users can perform tasks. Memorability — how easily users re-establish proficiency after not using the product. Errors — how often users make them, how severe they are, how easily users recover. Satisfaction — how pleasant the design is to use.
Usability is often used interchangeably with "UX", but they're different. UX is the whole experience; usability is one of its load-bearing dimensions. A beautifully designed product can have poor usability; a highly usable product can still produce a mediocre UX if desirability, trust or utility fail.
Products with poor usability lose users to abandonment, errors, and frustration — often silently, because users don't report "the interface was unusable", they just leave. ISO 9241-210 explicitly ties usability to business outcomes: completion rates, support costs, training costs, and task performance.
Usability is measured during evaluative research — usability testing, heuristic review, benchmarking studies. It's designed for during generative work — IA, interaction design, content design. The method finder → evaluate surfaces methods that measure it.
The three canonical usability measurements are task success rate, time on task, and SUS score. Together they cover effectiveness (success), efficiency (time), and satisfaction (SUS).
A usable site isn't necessarily the one with the fewest features — it's the one where the primary journeys complete reliably. Many feature-heavy enterprise products are high in capability but low in usability because the features aren't learnable, efficient or memorable.
Usability sits at the centre of UX measurement. It intersects with cognitive load (lower load typically means higher usability), affordance and signifier (visible cues drive learnability), and accessibility (the two overlap deeply but aren't identical).
Reviewed 1 October 2026 · Editorial standards