Skip to main content

Research Planning

Good research starts with good planning. This guide helps you scope research projects that answer the right questions within your constraints.

Last updated: February 2025

Why planning matters​

Research without planning tends to produce:

  • Vague findings that don't inform decisions
  • More questions than answers
  • Wasted time and participant goodwill
  • Results that arrive too late to matter

Planning isn't bureaucracy—it's thinking clearly about what you need to learn before you start trying to learn it.

The research planning process​

Step 1: Define the research question​

What do you need to learn? Be specific.

Vague: "Understand how users feel about our product" Specific: "Identify the top 3 barriers preventing new users from completing their first purchase"

Good research questions are:

  • Answerable through research (not opinion debates)
  • Specific enough to guide method choices
  • Relevant to decisions you need to make
  • Scoped to available time and resources
The decision test

Ask: "What decision will this research inform?" If you can't answer that, you're not ready to do research yet. Clarify the decision first.

Step 2: Identify what you already know​

Before gathering new data:

  • Review existing research (analytics, support tickets, past studies)
  • Talk to colleagues who interact with users
  • Document assumptions the team is making

Often, you'll find you have more information than you thought. Your research can then fill specific gaps rather than covering ground you've already covered.

Step 3: Choose your method​

Different questions require different methods:

Behavioral questions (what do people do?): Analytics, observation, usability testing

Attitudinal questions (what do people think/feel?): Interviews, surveys, diary studies

Qualitative questions (why? how?): Interviews, contextual inquiry, ethnography

Quantitative questions (how many? how often?): Surveys, analytics, A/B testing

Most research projects combine methods. A typical pattern:

  1. Qualitative research to explore and understand
  2. Quantitative research to measure and validate

Step 4: Define participants​

Who should you talk to? Consider:

Relevance: Do they represent your actual users?

Accessibility: Can you recruit them within your timeline?

Diversity: Are you including varied perspectives?

Sample size: Quality over quantity for qualitative; statistical significance for quantitative

For qualitative research (interviews, usability tests), 5-8 participants per user segment often reveals major patterns. You'll see diminishing returns after hearing the same themes repeatedly.

Step 5: Plan the logistics​

Before starting:

  • How will you recruit participants?
  • What incentive (if any) will you offer?
  • Where will sessions happen (in-person, remote)?
  • Who will facilitate? Who will take notes?
  • How will you record and store data?
  • What's your timeline?

Build in buffer time. Recruitment always takes longer than expected.

Step 6: Create your research plan document​

Document your plan for alignment and reference:

Research Plan: [Project Name]

Research Question: [Specific question]

Background: [Context and what you already know]

Method: [Approach]

Participants: [Who and how many]

Timeline: [Key dates]

Team: [Roles]

Deliverables: [What you'll produce]

Decision: [What this research will inform]

Scoping to constraints​

Limited time​

If you have days, not weeks:

  • Narrow the question to the most critical unknown
  • Use guerrilla methods (intercept testing, quick remote sessions)
  • Leverage existing data more, collect new data less
  • Focus on qualitative insights over quantitative validation

Limited budget​

If you can't offer incentives or hire recruiters:

  • Recruit through your product (in-app invitations)
  • Tap internal networks carefully (colleagues who match user profiles)
  • Use unmoderated remote testing tools
  • Focus on observation and analytics

Limited access to users​

If users are hard to reach:

  • Maximize each interaction (longer sessions, multiple methods)
  • Use proxy participants carefully (not ideal, but sometimes necessary)
  • Lean more heavily on behavioral data you can collect without direct contact
  • Build ongoing research relationships

Common research planning mistakes​

Trying to answer everything​

One study can't answer all your questions. Trying makes it answer none of them well. Pick the most important question and answer it properly.

Planning in isolation​

Research plans should involve stakeholders. Otherwise, you risk answering questions nobody cares about or missing questions that really matter.

Underestimating recruitment​

Recruitment is the most common schedule buster. People don't respond, they cancel, they no-show. Plan for 50% more outreach than you think you need.

Skipping the pilot​

Running one pilot session reveals problems with your protocol before you've burned through participants. Always pilot.

Over-documenting​

A 50-page research plan that nobody reads helps nobody. Keep plans concise and focused on what people actually need to know.

Working with stakeholders​

Getting buy-in for research​

Frame research in terms of risk reduction. "We're not sure if users will understand this navigation. Testing it takes 2 days and might prevent us from building the wrong thing for 2 months."

Managing expectations​

Be clear about what research can and can't deliver:

  • Research informs decisions; it doesn't make them
  • Research reveals patterns; it doesn't guarantee universality
  • Research takes time; rushing it produces unreliable results

Involving stakeholders appropriately​

Stakeholder involvement helps with buy-in and insight, but too much involvement creates problems:

Involve stakeholders in: Defining questions, reviewing plans, observing sessions, interpreting findings

Protect research from: Leading questions, confirmation bias, jumping to solutions during sessions

Practical templates​

Quick research brief (1 hour)​

When you need to move fast:

  1. What's the decision? (2 sentences)
  2. What do we need to know? (1 question)
  3. Who can tell us? (participant criteria)
  4. How many? (number)
  5. How will we learn? (method)
  6. When do we need answers? (date)

Full research plan (half day)​

For significant research:

  • Executive summary
  • Background and context
  • Research objectives and questions
  • Methodology
  • Participant criteria and recruitment
  • Timeline and milestones
  • Team and responsibilities
  • Risks and mitigation
  • Expected deliverables
How long should a research project take?

It depends on the question. Simple usability questions can be answered in a week. Foundational research into user needs might take months. Scope the research to the decision timeline—if you need an answer in two weeks, design research that fits.

How do I prioritize multiple research questions?

Rank by decision impact and urgency. What decision needs to be made soonest? What unknown creates the most risk? Research the intersection of important and urgent first.

When should I use surveys vs. interviews?

Surveys are good for measuring things you already understand. Interviews are good for learning things you don't yet understand. If you're not sure what questions to ask, start with interviews.

How do I know when I have enough data?

For qualitative research, saturation—when new sessions stop revealing new insights. For quantitative research, statistical significance. For practical purposes, when you have enough confidence to make the decision.

What if stakeholders want to skip research?

Understand why. Sometimes they're right—the question is already answered or the stakes don't justify the time. Sometimes they underestimate risk. Make the risk visible: 'If we guess wrong, we'll spend X months building something users don't need.'

Related resources