UX Research Plan Template: Questions, Methods, Participants and Decisions

A UX research plan template connects a product decision to an ethical, feasible evidence-gathering approach. It makes the questions, participants, methods, limitations and expected outputs explicit before recruitment or fieldwork begins.

The plan is not a guarantee of representative truth. Researchers and authorized product owners remain responsible for method quality, participant welfare, privacy, accessibility and interpretation. Follow the review processes that apply to the organization, population and data involved.

Begin with the decision, not the method

State the decision the research should inform, who owns it and when it must be made. Examples include choosing which workflow to improve, deciding whether a concept is understood or identifying barriers before a release. Avoid vague goals such as “learn about users.”

Document what is already known, the source and confidence. Separate assumptions from evidence. If leadership has already made an irreversible decision, be honest about whether the work is discovery, validation, risk reduction or communication; do not invite participation under a false premise.

Set boundaries for what the study will not decide. A usability interview cannot establish market size, legal compliance or clinical safety without the appropriate methods and qualified review.

Add a decision threshold where one can be defined responsibly. For example, specify which observed barriers would trigger redesign, further study or a limited pilot. Avoid inventing a numeric pass rate for a qualitative study merely to make the decision appear objective. Where no threshold is defensible, state who will weigh the evidence and which other inputs they must consider.

Identify stakeholders who may use the result and conflicts they bring. A product owner, designer, support lead and accessibility specialist may ask different questions of the same evidence. Agree which decision is primary, while preserving relevant findings that do not fit the immediate choice.

Write neutral research questions

Turn the decision into a small set of answerable questions about behavior, context, needs, comprehension or experience. “How do people currently complete this task?” is more useful than “Why do users prefer our faster solution?” because it does not embed an expected answer.

Distinguish research questions from interview questions. Research questions guide the study; session prompts gather evidence. Keep hypotheses visible where useful, but include plausible alternatives and evidence that could challenge them.

Use a consulting discovery questions structure for contextual inquiry, while removing assumptions that belong to a sales conversation. Research participants are not responsible for approving the team’s strategy.

Copy this UX research plan template

UX RESEARCH PLAN

Study title / owner / version:
Decision to inform / decision owner / date:
Background and verified evidence:
Assumptions and unknowns:
Out of scope:

RESEARCH QUESTIONS
1.
2.
3.
Evidence that could challenge the current view:

METHOD AND PARTICIPANTS
Method / rationale / limitations:
Target participants / relevant variation:
Inclusion / exclusion criteria:
Sample rationale / stop rule:
Recruitment source / incentive:

ETHICS AND DATA
Consent and withdrawal process:
Accessibility and safeguarding:
Data collected / purpose / access / retention:
Recording and transcription decision:
Required privacy, legal or ethics review:

FIELDWORK AND ANALYSIS
Session structure / pilot:
Moderator / observer roles:
Analysis method / evidence traceability:
Conflicts, deviations and missing data:

OUTPUT
Deliverable / audience / date:
Limitations to state:
Decision meeting / accountable owner:
Repository / deletion date:

Treat this as a control document. Update the version when scope or method changes and preserve material deviations rather than rewriting the plan as though they never happened.

Select a method that fits the question

Choose interviews for context and interpretation, observation for behavior in context, usability evaluation for interaction problems, diary approaches for change over time, and structured measurement when the decision needs quantification. Mixed methods may help, but only when each component has a clear purpose.

Explain why the method can answer the question and what it cannot establish. A concept testing interview guide can support early comprehension and relevance questions, but favorable reactions do not prove future adoption.

Check feasibility: access to participants, moderator capability, tools, language, accessibility, budget and decision timing. If constraints materially weaken the study, narrow the claim rather than disguising the limitation.

Document why plausible alternatives were not selected. Remote moderated sessions, unmoderated tasks and field observation create different visibility, participant burdens and technical risks. The cheapest or fastest method may still be appropriate, but the plan should show what evidence will remain unavailable. If the study changes method after fieldwork begins, record which sessions are comparable and which require separate interpretation.

Define participants and sampling rationale

Describe the experience or context participants need, then identify variation relevant to the decision: role, workflow, accessibility needs, device, geography or experience level. Do not collect demographic attributes without a justified purpose and appropriate handling.

State recruitment sources, inclusion and exclusion criteria, sample rationale, incentive and stop rule. There is no universal participant count. Qualitative work may seek sufficient range and thematic depth; quantitative work requires a design appropriate to the intended inference.

Avoid recruiting only convenient internal colleagues when the decision concerns external users. If convenience sampling is unavoidable, disclose it. A focus group discussion guide may help with shared perceptions, but group dynamics can suppress dissent and expose participants to one another.

Plan consent, privacy and participant care

Explain the study purpose, activities, duration, data use, recording, foreseeable discomfort, incentive, withdrawal route and contact information in understandable language. Consent should be informed and voluntary, not buried in a long document.

Collect only necessary data. Define storage, access, transfer, retention and deletion before sessions. Avoid exposing prototypes, participant identities or confidential business information through unapproved tools. Apply GDPR by design principles where relevant, while relying on qualified review for actual legal obligations.

Plan for accessibility, safeguarding and vulnerable populations. Researchers should know when to pause a session, decline a request or escalate a concern.

Design the session and pilot it

Build a session flow with welcome, consent confirmation, context, core activities, debrief and withdrawal reminder. Write neutral prompts and follow-ups. Avoid explaining the intended design before participants have had a chance to interpret it.

Define moderator and observer roles. Observers should not interrupt, contact participants independently or collect side-channel personal data. Establish how questions will be passed to the moderator and how confidential notes are handled.

Pilot the materials with someone outside the immediate design team when possible. Check timing, task clarity, technical setup, accessibility and whether prompts accidentally reveal expected answers. Record changes to the plan.

Prepare a disruption protocol. State what happens if the prototype fails, a participant cannot access a task, an observer reveals confidential context or a session surfaces distress. Define when to pause, reschedule, compensate and exclude compromised evidence. A failed setup can itself reveal accessibility or operational risk, but it should not be reframed as participant error.

Make analysis traceable to evidence

Choose the analysis approach before fieldwork: structured notes, thematic coding, task-level issue analysis or another defensible method. Define how conflicting observations, outliers and missing sessions will be handled. Keep a route back from each finding to authorized source evidence.

Separate observation, participant interpretation and researcher inference. A memorable quote is not automatically a dominant pattern. Report frequency cautiously in small qualitative samples and avoid percentages that imply population precision.

Plan how the team will reconcile coding differences and negative cases. Independent review can reveal assumptions, but agreement between researchers does not make an interpretation objectively true. Record material disagreements and the basis for the final synthesis. If sessions occur in multiple languages, document who translated or interpreted them and where meaning may have shifted. Preserve enough context for a responsible reviewer to challenge the analysis without granting broad access to participant data.

Use a decision log template to connect findings to product choices, alternatives, uncertainty and owners. The decision owner may choose differently for strategic reasons, but the record should not rewrite the evidence.

Capture sessions responsibly

Recording should be necessary for the stated purpose, covered by the approved consent process and limited to authorized people and systems. Provide an alternative or stop recording if required. Do not assume that agreeing to participate means agreeing to every secondary use.

For authorized research with explicit participant awareness, Kuno can help create draft transcripts, notes and action items for researcher verification. It does not obtain consent, anonymize by default or determine findings. Explore Kuno

Check transcripts against recordings where accuracy matters, remove unnecessary identifiers and preserve uncertainty. Automated summaries can omit dissent or context, so researchers must review source material.

Report limitations and support the decision

Structure the output around the decision and research questions. Include method, participants, evidence, findings, contradictions, limitations and recommendations with owners. Distinguish an observed issue from a proposed design response.

Share participant data only with authorized audiences. Use a content brief template pattern to align audience, purpose and evidence, but do not flatten nuanced research into unsupported marketing claims.

After the decision, archive or delete data according to the plan and review whether the study answered the question. Capture method improvements separately from findings.

Keep authorized research conversations traceable without outsourcing interpretation. Kuno supports draft transcripts and summaries while researchers verify evidence, privacy, limitations and decisions. See Kuno

FAQ

The FAQ below covers plan contents, research questions, participant numbers, approvals and responsible use of AI in UX research.