Skip to content
UI/UX15 min read

User Research for Product Design: Methods, Process and Best Practices

How to plan and run user research: choosing between interviews, surveys, analytics, usability testing and more, recruiting participants, and turning findings into design decisions.

01

Quick answer

User research is the systematic study of users' goals, behaviors and pain points to guide design decisions. Choose methods by the question: interviews and observation to understand why people act as they do, surveys and analytics to measure how many and how often, and usability testing to check whether a design works. Recruit participants who match real users, keep studies small and frequent, synthesize findings into themes, and share them so product decisions rest on evidence rather than opinion.

02

Why User Research Matters

Teams are rarely representative of their users. Research exposes the gap between what the team assumes and what users actually need, before that gap becomes expensive in engineering time. It's the foundation of the product design process.

03

Types of Research

Two dimensions help choose methods: attitudinal (what people say) vs behavioral (what they do), and qualitative (why and how) vs quantitative (how many and how much). Nielsen Norman Group's guide to choosing UX research methods maps many methods on these dimensions.

04

Common Methods

MethodBest forType
User interviewsGoals, motivations, context, pain pointsQualitative, attitudinal
Contextual inquiryHow people work in their real environmentQualitative, behavioral
SurveysMeasuring attitudes across many usersQuantitative, attitudinal
Analytics reviewWhat users do at scale, where they drop offQuantitative, behavioral
Usability testingWhether a design works for real tasksQualitative, behavioral
Card sortingHow users group and label contentQualitative or quantitative
Tree testingWhether users can find items in a structureQuantitative, behavioral
Diary studiesBehavior over days or weeksQualitative, behavioral
A/B testingWhich version performs better at scaleQuantitative, behavioral
05

Research Across the Product Lifecycle

Discovery research, through interviews, observation and analytics, shapes what to build. Evaluative research, such as usability tests and tree tests, checks designs before launch. Post-launch research, such as analytics, surveys and support analysis, shows what to improve. Mature teams run all three continuously.

06

Planning a Study

  • Write the decision the research should inform
  • List the specific questions to answer
  • Choose methods that fit those questions
  • Define who to recruit and how to screen
  • Prepare a discussion guide or task list
  • Decide how findings will be shared
07

Recruiting Participants

Recruit people who match your real users' behaviors, not just demographics. Existing customers, waitlists and recruitment panels are common sources. Screen out colleagues and friends, and compensate participants fairly for their time.

Need research before your next big design decision?

ZSpace Labs plans and runs user research, from interviews to usability tests, and turns findings into design direction.

Start a Project
08

Running Interviews

Ask about past behavior ("Tell me about the last time you...") rather than hypothetical preferences. Use open questions, follow up on specifics, and avoid leading questions or pitching your idea. Record, with consent, so you can listen rather than take notes.

09

Synthesis: From Notes to Insights

Group observations into themes (affinity mapping), look for patterns across participants, and write insights that link a behavior to its cause and a design implication. Personas and journey maps are useful ways to share these findings, as long as they're grounded in the research.

10

Research Ethics and Privacy

Get informed consent, explain how recordings and data will be used, store them securely, and minimize personal data. Research data is personal data, and privacy obligations vary by jurisdiction.

11

Common Mistakes

  • Asking users what features they want instead of observing their problems
  • Leading questions that confirm existing ideas
  • Recruiting people who don't match real users
  • Relying only on surveys or only on analytics
  • Research reports nobody reads or acts on

Want research built into your product process?

Talk to ZSpace Labs about UI/UX design grounded in real user evidence.

Start a Project
12

Research for B2B and Complex Workflow Software

B2B products have several roles in one customer: the buyer, administrators, daily users and approvers often want different things. Research needs to cover each role and the workflow between them, not just the person who signed the contract.

  • Recruit through customer success and account managers, with the customer's permission.
  • Observe real work in context (contextual inquiry) where possible; workarounds and spreadsheets reveal unmet needs.
  • Map the workflow across roles and systems before designing screens.
  • Respect confidentiality: agree what you can record, use NDAs where needed, and de-identify notes.
  • Expect small samples; combine interviews with product analytics and support data.
  • Test with realistic data and permissions, not empty demo accounts.
13

Conclusion

Good user research answers specific questions with the right methods, from representative people, and turns findings into decisions the whole team understands. Start small with analytics and a few interviews, then make research a habit. Next, see usability testing and the product design guide. For a government model of research-led delivery, see the GDS Service Standard for product teams.

For related guides, see ecommerce conversion research.

FAQ

Common questions.

Systematic study of users' needs, behaviors and motivations to inform design decisions, using methods such as interviews, observation, surveys, analytics and usability testing.

Qualitative research (interviews, usability tests) explains why people behave as they do, usually with small samples. Quantitative research (analytics, surveys) measures how many and how often, with larger samples.

Attitudinal research captures what people say (interviews, surveys). Behavioral research captures what they do (analytics, usability tests). What people say and do often differ.

Often five to eight per distinct user group is enough to see recurring patterns. Continue until you stop hearing new themes.

Throughout: discovery research before defining the problem, evaluative research during design, and ongoing research after launch.

From existing customers, sign-up lists, recruitment panels or targeted outreach, screening for the behaviors and characteristics of your actual users.

Yes. Analytics review, a handful of remote interviews and quick unmoderated tests can deliver valuable insight at low cost.

Summaries of research about distinct user groups, describing their goals, contexts and pain points. They should be based on research, not invented.

Synthesize findings into themes, prioritize problems by impact, write clear problem statements, and share them with the whole team so decisions reference evidence.

B2B research has to cover several roles per customer (buyers, administrators, daily users, approvers), complex workflows across systems, and confidentiality constraints. Recruit through account teams with permission, observe real work in context, map workflows across roles, and combine small-sample interviews with analytics and support data.

Get in touch

Have a project in mind?

Whether you're building a new digital product, improving an existing website, or looking to automate part of your business — let's talk.