How to Run a Focus Group From Scratch

How to Run a Focus Group From Scratch

6 minute read

Most organizations that want to run a focus group either over-engineer it (hiring an expensive research firm for a question that five customer conversations would answer) or under-engineer it (putting ten people in a conference room with a list of questions and hoping for insight). Neither approach produces reliable findings. Here's how to do it right.

This is a step-by-step guide for running your first focus group. For expert facilitation and recruitment support, see our focus groups service page — but if you're doing it yourself, this guide covers everything you need.

We'll cover:

• What a focus group is actually for

• The 6-step focus group process

• How to write a discussion guide

• How to recruit participants

• How to facilitate (and what to avoid)

• How to analyze and report findings

• Frequently asked questions

Table of Contents

1. 1. What a focus group is actually for
2. 2. The 6-step process
3. 3. How to write a discussion guide
4. 4. How to recruit participants
5. 5. How to facilitate
6. 6. How to analyze and report
7. 7. Frequently asked questions
8. 8. Key tips

1. What a Focus Group Is Actually For

A focus group is a qualitative research method that brings a small group of participants together to discuss a topic in depth. The goal is to surface language, reveal reactions, and understand the reasoning behind opinions — not to count how many people hold a particular view.

Focus groups are excellent for understanding how people think and talk about a problem, for revealing unexpected concerns or objections, and for testing concepts and messaging before you invest in production. They're poor predictors of behavior and shouldn't be used to estimate how many people will buy something.

For a deeper look at when focus groups are the right method vs. alternatives, see our market research methods guide.

2. The 6-Step Focus Group Process

Step 1: Define your research question.

Before anything else, write the one question this focus group needs to answer. 'How do small business owners describe the problem of managing cash flow?' is a research question. 'What do customers think about our brand?' is not — it's too broad to produce useful findings.

Step 2: Design your screener.

A screener is a short questionnaire that identifies participants who match your target profile. It should take under three minutes to complete and include three to five must-have qualification criteria.

Step 3: Recruit participants.

Six to ten participants per group is the right range. See our dedicated guide on how to recruit focus group participants for sourcing, screening, and incentive guidance.

Step 4: Write your discussion guide.

A discussion guide is an outline of your questions and probes, not a rigid script. See Section 3 for how to write one.

Step 5: Facilitate the session.

Run the session (in person or via Zoom) with a facilitator who isn't emotionally invested in the outcome. Record with participant consent.

Step 6: Analyze and report.

Review the recording or transcript. Identify recurring themes. Write up findings in plain language. See Section 6.

3. How to Write a Discussion Guide

A focus group discussion guide is a flexible outline, not a questionnaire. The difference matters: a questionnaire expects specific answers; a discussion guide creates space for unexpected answers.

Structure:

• Opening (15 min): Welcome, ground rules, warm-up question. 'Tell us your name and one word that describes how you feel about [topic].'

• Core questions (45–55 min): 3 to 5 open-ended questions that address your research question directly. Each should be followed by probes: 'Can you say more about that?' 'What do you mean by X?' 'Has anyone had a different experience?'

• Closing (10 min): 'Is there anything important we haven't discussed?' 'What's the one thing you'd want us to know?' Give every participant a final chance to speak.

Question writing rules:

• Start broad, move to specific

• Use open-ended questions only (no yes/no questions)

• Never embed your hypothesis in the question: 'Why do you think X is a problem?' assumes X is a problem

• Keep each question under 25 words

A focus group isn't a survey read aloud. It's a structured conversation with room for the unexpected. Design for what you don't know, not just what you think you know.

4. How to Recruit Participants

According to the Insights Association's standards for focus group research, using poorly screened participants is the most common cause of unreliable focus group findings. Getting recruitment right is as important as any other step in the process.

• Define your must-have criteria. What characteristics are absolutely required for this person to be a useful participant? These become your screener questions.

• Over-recruit by 20 percent. For an 8-person group, recruit 10. Last-minute dropouts are normal.

• Pay appropriate incentives. General consumer participants: $50 to $100 for a 90-minute session. Professional or specialized audiences: $150 to $300+.

• Exclude known biases. Employees of competing companies, market researchers, and people with close relationships to your organization should all be screened out.

5. How to Facilitate

The facilitator's job is to create conditions for honest, productive conversation — not to confirm your hypotheses. The most important facilitation rule: stay neutral. Your reaction to participant responses (verbal or non-verbal) shapes what they say next.

Techniques that work:

• The 5-second pause. After a participant answers, wait 5 seconds before responding. More often than not, someone will add something important.

• The go-round. 'Let's hear from everyone on this — starting on the left.' This prevents dominant voices from shaping the entire discussion.

• The reflect-and-probe. 'You said X — can you say more about what you mean by that?' This is your primary tool for depth.

• The temperature check. 'We've heard a few different perspectives here. How many people share [name]'s view?' A show of hands gives you a quick sense of majority vs. minority positions.

6. How to Analyze and Report Findings

After each session, write a brief summary of the top three to five themes while the session is still fresh. Then review the full transcript or recording for depth. According to the American Psychological Association's guidelines on qualitative research reporting, rigorous qualitative analysis requires systematic data review, not impressionistic summary. Take the analysis seriously.

• Identify themes, not just quotes. Look for patterns across participants, not just memorable individual statements.

• Note what was surprising. The findings that challenge your assumptions are usually the most valuable.

• Separate reaction from rationalization. What people do first (lean in, laugh, go quiet) tells you as much as what they say afterward.

For guidance on how to package your findings, see our post on research reports vs. insight briefs.

Frequently Asked Questions About Running a Focus Group

How many focus groups do I need?

Two to three groups minimum before drawing conclusions. One group is an anecdote. Two groups with consistent themes start to become a finding. Three groups producing the same themes are a credible pattern.

Can I facilitate my own focus group?

You can, with caveats. If you're the product creator or the business owner, participants often soften their negative reactions to protect your feelings, even when they're trying to be honest. An external facilitator typically produces more candid data. If you do facilitate, ask a colleague to take notes so you can focus on listening.

Virtual vs. in-person — which is better?

Both work. Virtual focus groups (Zoom or similar) are faster to schedule, eliminate geographic constraints, and often produce comparable quality to in-person sessions for most topics. In-person remains preferable for topics involving physical products, sensory experience, or populations with lower technology comfort.

Key Tips for Running Your First Focus Group

• Write your research question before you write a single discussion guide question.

• Over-recruit by 20 percent. Assume someone won't show.

• Record everything with consent. Memory is unreliable; recordings are not.

• Run two to three groups before drawing conclusions.

• Stay neutral as a facilitator. Your reactions shape participant responses.

How Praxia Insights can help

At Praxia Insights, we design and run research that gets to the real answers. Whether you need focus group facilitation, a polished insight brief, or a full research plan built from scratch, we're here for it.

Schedule a Consultation