Research Brief Template: How to Define Your Research Before You Start

A research brief is the single most important document you'll create before starting any user research project. It defines what you're trying to learn, who you're learning it from, and why it matters — keeping everyone aligned before a single interview is conducted.

Without a brief, research drifts. Interviews wander. Analysis takes forever because you're not sure what you were looking for in the first place. With a strong brief, every interview feels purposeful and the insights practically write themselves.

What Is a Research Brief?

A research brief is a structured planning document that answers five core questions:

1. What decision does this research need to inform?
2. What do we already know, and what's the gap?
3. Who is the right participant to talk to?
4. What methodology will we use?
5. What will we do with the findings?

Think of it like a project charter for research. Short enough to share in a Slack message but comprehensive enough to guide a month of fieldwork. According to the Nielsen Norman Group, research that starts with a well-defined brief produces insights 3x more likely to influence product decisions. When stakeholders can see exactly what research is designed to answer, they engage with the findings rather than questioning the methodology.

The Research Brief Template

Here's a proven template you can adapt for any user research project — UX research, customer discovery, usability testing, market research, or employee experience research.

---

Section 1: Research Context

Problem Statement What problem are you investigating? Write 2-3 sentences describing the situation from the user's perspective — not a feature request or business metric.

Example: "New users who sign up for our product don't activate within the first 7 days. We believe this is a friction problem in onboarding, but we don't know which steps cause drop-off or why."

Decision to Inform What specific decision will this research inform? Vague research questions lead to vague findings.

Example: "This research will inform our Q3 onboarding redesign. We need to decide whether to simplify the flow (fewer steps) or improve it (better guidance at each step)."

Hypothesis What do you believe is true, and why? Writing your hypothesis down helps you test it rather than unconsciously confirm it.

Example: "We believe users abandon onboarding because the initial setup requires too many manual inputs before they see any value."

---

Section 2: Research Objectives

List 3-5 specific things you want to learn. Use "understand" language, not "validate" language.

• Understand how users describe the problem in their own words
• Understand what success looks like for the user at this stage
• Understand what context users bring to the experience (prior tools, mental models)
• Understand what friction points cause users to pause or abandon

Pro tip: If you catch yourself writing "validate that users want X," rewrite it as "understand whether X actually solves the user's problem."

---

Section 3: Target Participant Profile

Who should we talk to? Define your ideal participant with specific behavioral or situational criteria — not demographic ones.

Example: "Users who signed up in the last 30 days and have not completed their profile setup."

What experience should they have? What does the participant need to have done or experienced to give you useful insights?

Example: "Must have attempted to set up the integration at least once. Does not need to have succeeded."

Screening Question The one question that separates the right participants from the wrong ones.

Example: "Have you tried to connect a third-party tool to our product in the last 30 days?"

How many participants? For qualitative research, 5-8 participants per distinct segment is usually sufficient. According to Nielsen Norman Group, 5 users uncover approximately 85% of usability issues in any given interface. For broader discovery studies, 8-12 gives you higher confidence in theme frequency.

---

Section 4: Methodology

Choose an approach based on what you need to learn:

For each methodology, define your question format:

• Core questions: The 4-6 main questions you'll always ask
• Follow-up probes: Specific "tell me more" follow-ups for each core question
• Topics to avoid: Questions that bias responses or break rapport

---

Section 5: Success Criteria

Define success before you start — not after.

• We have 5+ completed interviews with the target profile
• We can articulate the top 3 friction points users face
• We have at least 2 compelling user quotes per key finding
• We can answer the original decision question with confidence

Distribution plan: Who receives the report? In what format? By when? If you can't answer this, the research isn't ready to start.

---

How AI Tools Like Koji Generate Research Briefs Automatically

Traditionally, creating a research brief like this takes 2-4 hours of facilitated workshops with PMs, designers, and researchers debating scope and objectives.

Platforms like Koji automate this entirely. When you describe your research goal in a few sentences, Koji's AI consultant asks clarifying questions and generates a complete, structured research brief — including problem context, participant profile, methodology choice, and interview questions — in under 10 minutes.

The brief isn't static either. Koji's AI adapts it as you learn more, automatically incorporating context files (competitive analyses, product specs, prior research) into the interview design. The result feels custom-designed by an experienced researcher, ready to launch immediately. Instead of a workshop, you have a conversation.

Common Research Brief Mistakes

Mistake 1: Writing objectives as features "Understand if users want a dark mode" is a feature question disguised as research. Better: "Understand how users manage screen fatigue during long work sessions."

Mistake 2: Recruiting by demographics instead of behaviors Age and job title are poor predictors of relevant insights. Recruit by what people have done or experienced.

Mistake 3: Too many objectives If your brief has 12 research objectives, it has zero. Pick 3-5 and be ruthless. Deep on three beats shallow on twelve every time.

Mistake 4: No decision link Research that doesn't connect to a specific decision rarely gets acted on. Every brief should start with: "This research will help us decide..."

Mistake 5: Skipping the hypothesis Not writing your hypothesis doesn't make research more objective — it makes it more susceptible to post-hoc rationalization. Write it down, then try to disprove it.

Using Your Brief to Guide Analysis

A well-written research brief makes analysis dramatically faster. After your interviews, compare what you heard against each objective:

1. Did you answer all 3-5 research objectives? If not, why not?
2. Was your hypothesis confirmed, disconfirmed, or complicated by what you found?
3. Which insights directly inform the decision you started with?

This is how you go from "we talked to 8 people and it was interesting" to "we have three findings that will change our roadmap." The brief is your analysis scaffold.

Research Brief Template Checklist

Before starting any research, verify your brief includes:

• A problem statement describing user experience, not business metrics
• A specific decision this research will inform
• A written hypothesis
• 3-5 research objectives (not more)
• A participant profile based on behavior, not demographics
• A screening question
• A methodology choice with rationale
• Success criteria defined in advance
• A plan for sharing and acting on findings

Tips & Best Practices

• Share the brief before the study begins — alignment before research prevents scope creep and stakeholder disputes about findings later
• Keep it to 1-2 pages — a brief longer than that is trying to answer too many questions at once
• Update it if early interviews reveal wrong assumptions — a living brief is better than a rigid one that leads you astray
• Use it as your analysis scaffold — every insight should trace back to a brief objective
• Let Koji generate your first draft — describe your research goal and the AI consultant creates a complete brief you can edit, not a blank page

Frequently Asked Questions

How long should a research brief be? One to two pages maximum. A brief is a planning document, not a research proposal. If it's longer, you're trying to answer too many questions at once.

Who should approve the research brief? The person who will act on the findings (usually a PM or design lead) and anyone conducting the interviews. Everyone should agree on the objectives before starting.

Can I change my brief mid-research? Yes — and you should if your first few interviews reveal your original objectives were off. Updating the brief mid-study is better than forcing irrelevant questions on every subsequent participant.

How is a research brief different from a discussion guide? A brief is the strategic "why and what" — it defines what you're trying to learn. A discussion guide is the tactical "how" — the specific questions you'll ask. Your brief informs your discussion guide.

Does Koji create a research brief automatically? Yes. When you describe your research goal to Koji's AI consultant, it generates a complete structured brief including methodology, participant criteria, and interview questions. You can edit any section manually or refine it through conversation. Most users have a publishable brief in under 10 minutes.