Generative research
Generative research, also called discovery or exploratory research, aims to understand people's needs, behaviors and contexts in order to find problems worth solving, before any solution exists. It generates the questions and opportunities that later research evaluates.
How it works
Generative studies are open-ended. Typical methods include in-depth interviews, field studies, diary studies and ethnography, asking about goals, routines, frustrations and workarounds rather than reactions to a design. The output is understanding: problem statements, opportunity areas, personas, journey maps and hypotheses.
Nielsen Norman Group describes discovery as the phase that explores the problem space without testing a hypothesis or a solution. It contrasts with evaluative research, which tests something that exists. It is most useful at the start of a product, a market entry or a strategy review, when the options are widest, and when an established product stalls for reasons the metrics do not show. Its findings set the brief for the evaluative research that follows.
A worked example
A health insurer plans a wellness app. Generative interviews with 24 members, before any design, find that 17 care less about exercise tracking than about understanding their bills and coverage. The first release becomes a plain-language coverage explainer, and the step counter is shelved.
Related
Automate this in Pulse
Generative research can start before there is a customer to ask, and Pulse can recruit to a profile built around the activity under study, such as people planning a house move in the next year.