User Research Methods

Replace assumptions with evidence: match research type to the problem's maturity, treat research as ethnography, research the organization and stakeholders as rigorously as users, name and manage bias, and build personas from firsthand research.

When this applies

• The goal is content or feature brainstorming with target audiences (P002).
• Moving from IA research toward strategy or defending research effort (P009).
• Prioritizing components, labels, navigation, search, or content access for a product (P018).
• choosing a research activity for the current project stage (P038).
• planning research into the people a design serves (P039).
• Starting an IA project or redesign (P046).
• Researching user needs, language, behavior, or satisfaction under practical constraints (P047).
• Planning or running the transition from IA research to strategy (P048).
• starting work inside or for an organization (P056).
• designing, running, or interpreting any study (P057).
• structuring any research effort (P058).
• running an analysis session (P059).
• writing a screener or selecting participants (P060).
• creating or using personas to guide design (P061).

Procedure

Apply these principles to the situation under review; for each, name the user need at stake and the trade-off the choice carries.

1. Use focus groups for early planning or ideation, not as the primary way to validate IA usability or architecture decisions. (P002)
2. Treat research and strategy as connected work: the same team should carry hands-on learning into strategy, record design ideas during research, and use IA-specific research to counter unsupported assumptions. (P009)
3. Start IA decisions from real users and their information needs, using field or behavioral research rather than internal brainstorming alone. (P018)
4. Match the type and openness of research to the maturity of the problem: use generative or exploratory research to define an ill-defined problem, descriptive research to understand context once a problem is identified, evaluative research iteratively to test solutions, and causal research to explain a live product, and do not fine-tune solutions while the problem is still undefined. (P038)
5. Do user research to replace assumptions with patterns and genuine empathy, treating user research as ethnography that understands how and why people behave in their real context rather than gathering opinions through surveys or focus groups, and observe representative users in their own environment across physical setting, mental model, habits, and relationships. (P039)
6. Begin context research with background materials, a shared IA kickoff, focused research meetings, early IT engagement, and stakeholder interviews. (P046)
7. Use multiple lightweight user-research approaches, because any real user evidence is better than none and different methods reveal different parts of the problem. (P047)
8. Start strategy hypotheses before research, test them during research, and shift from discovery to strategy when learning plateaus and candidate structures need validation. (P048)
9. Research the organization as rigorously as its users: get out and talk to real people, interview the stakeholders whose support the project needs (anyone whose lack of support would sink it) using user-research methods, and gather clear business requirements, because work that does not support the business fails no matter how good the design. (P056)
10. Name and manage bias in every study: assume bias is unavoidable and note it so you can weight results appropriately using a bias checklist, keep interviewers neutral especially early, reduce sponsor bias by not naming the sponsor prematurely, counter social-desirability bias by promising confidentiality and stressing honesty, and counter the Hawthorne effect by blending in. (P057)
11. Follow a lightweight but explicit process: define the problem, select the approach (its type from the problem statement and its specific method from available resources), plan and prepare by naming a point person and sketching a revisable plan with tradeoffs and fallbacks decided in advance and roles, recruiting, and materials listed, then collect, analyze, and report. (P058)
12. Analyze collaboratively and structurally: involve as many team members as possible, follow an explicit analysis structure even for a two-person team, give participants advance access, look for goals, priorities, tasks, motivators, barriers, habits, relationships, tools, and environment, turn patterns into observations and then recommendations against the original problem statement, and group around patterns that emerge rather than ones imposed in advance. (P059)
13. Recruit deliberately for representativeness: screen primarily for the behaviors that matter (a good participant shares the target's goals and key characteristics, can articulate their thoughts, and matches the target's technology familiarity), use a short screener that filters out time-wasters and stays vague about the test, verify tool and domain knowledge realistically, and phone-follow-up in-person candidates. (P060)
14. Build and use personas as the users in user-centered design: create composites from firsthand research, keep them as few as possible with multiple roles, brief and reusable, with realistic construction details, keep design targets separate from marketing targets, and make personas the first people you check any new idea against. (P061)

Principles applied

• P002 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P009 (medium) — grounded in claims/evidence and chunk anchors in sources/.
• P018 (medium) — grounded in claims/evidence and chunk anchors in sources/.
• P038 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P039 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P046 (medium) — grounded in claims/evidence and chunk anchors in sources/.
• P047 (medium) — grounded in claims/evidence and chunk anchors in sources/.
• P048 (medium) — grounded in claims/evidence and chunk anchors in sources/.
• P056 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P057 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P058 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P059 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P060 (high) — grounded in claims/evidence and chunk anchors in sources/.
• P061 (high) — grounded in claims/evidence and chunk anchors in sources/.

Provenance

Grounded in principles P002, P009, P018, P038, P039, P046, P047, P048, P056, P057, P058, P059, P060, P061, their backing claims and evidence records, and paragraph-level source anchors under sources/anchors/. Every cited id resolves into this package's distilled spine; see provenance- ledger.md and reports/faithfulness-report.yaml.