Choosing a method for the phase you are in, and how many people to test with.
AI workflow Find candidate themes in interview notes
Viewport: All
Viewport references are useful starting points. Design breakpoints should respond to the content, not specific device models. Choosing a viewport changes what is emphasised; nothing is hidden.
3 axes
Applies to All viewports
Attitudinal or behavioural: what people say or what they do. Qualitative or quantitative: observed directly or measured indirectly. Context of use: natural, scripted, limited or decontextualised.
Source
Applies to All viewports
Strategise, to find new directions: field studies, diary studies, interviews, surveys, concept testing. Design, to improve usability: card sorting, tree testing, usability testing. Launch and assess, to measure performance: benchmarking, A/B testing, analytics, surveys.
Source
5 users
Applies to All viewports
NN/g advises testing with no more than 5 users and running as many small tests as you can afford: three studies of 5, not one of 15. With distinct user groups, test 3 to 4 per group for two groups, or 3 per group for three or more.
Source