Definition
UX research is the systematic study of users — their behaviours, needs, motivations, and constraints — through interviews, observations, surveys, usability tests and analytics, used to inform product and design decisions with evidence rather than assumption.
UX research is the discipline that separates opinion-driven design from evidence-driven design. Without it, the loudest stakeholder usually wins design arguments — and the user pays the price. With it, the team has a shared source of truth to reference instead of relitigating the same debates every sprint.
Modern UX research splits into two complementary modes: generative (discovery work that uncovers needs and opportunities) and evaluative (testing whether a specific design solves the problem it's meant to solve). Both have their place; teams that skip generative research end up solving the wrong problems beautifully.
Origin
Human factors research roots trace to WWII military equipment design. The contemporary UX research practice formalised through the 1990s and 2000s with foundational texts from Jakob Nielsen and Don Norman.
How it works
- Define the research question — what do we need to learn?
- Pick the right method (interview, survey, usability test, observation, diary study, etc.).
- Recruit 5-15 representative participants.
- Run sessions, capture verbatim observations.
- Synthesise patterns across sessions — what's said more than once across users?
- Translate insights into design implications, not just observations.
When to use it
Use when
- At the start of any new product or major feature.
- When the team disagrees about user behaviour.
- When the data shows usage patterns the team didn't predict.
Skip when
- For very small UI changes within established patterns.
- When the cost of research exceeds the cost of just shipping and observing.
Key metrics
- Sessions completed
- Insights documented
- % of design decisions traceable to research
- Build rework rate (lower with good research)
Examples
- Six user interviews caught a fundamental misconception we'd been designing around for two months.
- Our research said users wanted X; we shipped Y; we lost.
- Most expensive design mistakes get made when teams assume they know the user.
In practice at Makreate
Makreate UX engagements are research-led by default. Even when the brief calls for 'just design,' we typically run 5-8 user interviews in the first week to anchor the rest of the project in evidence. Clients who initially resist this approach almost always change their mind by week two — the research findings tend to redirect the work in ways that no amount of design instinct would have caught on its own.
UX Design →Common mistakes
- Treating research as a one-time discovery phase instead of an ongoing practice.
- Asking users what they want instead of observing what they do.
- Confirming bias by recruiting participants who already agree with the team.
- Running too few sessions to identify patterns.
- Not synthesising — research without synthesis is just notes.
Frequently asked
How many users do I need for UX research?
5-8 users surface 80% of issues in any single study. For generative discovery, 10-15. For statistically significant quantitative results, much more.
What's the difference between UX research and market research?
Market research studies who buys and why. UX research studies how people use products. Both are useful; they answer different questions.
Can I do UX research without a researcher?
Yes — for simple usability testing especially. For complex generative research, a trained researcher catches insights non-researchers miss.