Skip to content
TL;DR: What is a Focus Group?
A focus group is a moderated discussion where a small group of people (usually 6 to 10) react to a product, idea, or topic so you can hear the why behind their opinions. It’s qualitative research, so you get depth, statistics, and actionable insights.
This guide covers what focus groups are, the main types, and a six-step process for running one, from screening participants to analyzing the transcript (the part where recording tools end up saving you a few hours).
A focus group is a small, moderated group discussion, designed to explore how people think and feel about a specific product, service, or idea. Unlike a survey, which hands you percentages, a focus group hands you the reasoning behind them: the hesitations, the tangents, and the offhand comment that turns out to be the actual insight.
That reasoning is the whole point of focus groups. It’s why they have outlasted seventy years of research fads. The bad news is that a group discussion generates a wall of unstructured talk that someone has to sit through, transcribe, and make sense of. The good news is you no longer need a full-on research agency to do it — and the analysis part has gotten dramatically less painful.
Here’s how the method works, and how to run one without losing a week to your notes.
What Is a Focus Group?
A focus group is a research method in which a trained moderator leads a small group of carefully selected people through a guided discussion about a specific topic. The goal is to collect qualitative data (think: perceptions, attitudes, and motivations) by letting participants react to a subject and to each other, which surfaces opinions a one-on-one conversation often misses.
The word “focus” is the important thing. You’re not here for a cup of tea and a chat; it’s a structured discussion organized around a defined research question, with a question sequence the moderator follows. Foundational researchers Richard Krueger and Mary Anne Casey describe a focus group as a special kind of group defined by its purpose, size, composition, and procedures. It’s a discussion engineered to obtain perceptions in a permissive, non-threatening environment. In practice, that means a handful of strangers in a room (or on a call), a moderator with a script, and a recorder running in the background.
What Is the Purpose of a Focus Group?
The purpose of a focus group is to understand the reasoning, language, and emotion behind people’s opinions. In focus groups, you’re less interested in how many people hold the same opinion (though that can be useful), and more in why they hold the opinions they do.
Teams use them to test product concepts before a build, pressure-test marketing messages, explore why customers churn, and dig into pain points that numbers alone can’t explain. This is why product discovery workflows so often lean on them. Focus groups sit alongside methods like user interviews and usability testing in the qualitative toolkit, and they’re common everywhere from product development to political polling to healthcare research.
They’re especially useful early, before you’ve committed engineering time to an idea. A ninety-minute session can reveal that the feature you were about to build solves a problem nobody actually has. That’s a much cheaper lesson to learn in a conference room than in production.
Is a Focus Group Qualitative or Quantitative Research?
A focus group is qualitative research. It produces rich, descriptive data about why and how people think, expressed in their own words, rather than the countable, statistically representative data a survey delivers.
This distinction is key: it’s the reason you should never present focus group findings as percentages (“80% of participants preferred…”), because eight people are not a statistically valid sample of anything. Focus groups tell you what’s possible and worth exploring; surveys tell you how common it is.
Types of Focus Groups
There are several types of focus groups, distinguished mainly by whether they run in person or online, how large they are, and how the moderation is structured. The right format depends on your topic, budget, and how sensitive the subject is. If your focus group is discussing a new snack flavor, it has different needs than one discussing a healthcare experience.
Here are the formats you’ll actually encounter:
| Type | Format | Typical size | Best for |
|---|---|---|---|
| Traditional (in-person) | Face-to-face in a facility or room | 6–10 | Reading body language, testing physical products or packaging |
| Online (synchronous) | Live video call | 6–10 | Geographically spread participants, lower cost, easy recording |
| Online (asynchronous) | Discussion board over days | 10–30 | Busy participants, reflective topics, no scheduling clashes |
| Mini focus group | In-person or online | 3–5 | Complex or sensitive topics needing depth over breadth |
| Dual-moderator | Two moderators share roles | 6–10 | Technical topics where one leads and one probes |
| Two-way | One group observes another | 6–10 each | Studying how one group reacts to another's discussion |
| Sources: Krueger & Casey (Sage); Nyumba et al., Methods in Ecology and Evolution (2018) | |||
Online vs. In-Person Focus Groups
Online focus groups trade a little richness (subtle body language, the energy of a shared room) for a lot of practicality: no travel costs, national or global recruiting, and a discussion that’s recorded and transcribable by default.
In-person groups still win for physical products and highly sensitive topics where trust is built in the room, but for most teams running remote research, the online format is now the default. Even though a single Google Meet or Zoom call can host far more than ten people, you’ll want to keep the group small enough to stay a discussion. Plenty of teams already run their remote UX research this way.
How Many People Should Be in a Focus Group?
A focus group should have 6 to 10 participants. That range is the widely accepted sweet spot: large enough to generate a variety of perspectives and cross-talk, small enough that everyone gets airtime and the conversation doesn’t fracture into side chats. Research on focus group methodology finds that groups of roughly six to eight are sufficient for most purposes, and that once you push past twelve, the group tends to fragment and become hard to moderate.
When the topic is complex, sensitive, or emotionally charged, you’ll want an even smaller group. Mini groups of 3-5 tend to work better since fewer people means more depth per person and more psychological safety. Go toward the upper end when the topic is light and you want breadth of opinion.
One group is rarely enough, though. A single focus group tells you what one set of eight people thinks. If you want to spot insightful patterns, you’ll typically need to run three to five groups per audience segment and keep going until new sessions stop surfacing new themes. That’s the qualitative version of a big enough sample: you’re looking for the point where you’ve heard it all before.
How to Run a Focus Group (Step by Step)
To run a focus group, define your objective, recruit and screen the right participants, write a discussion guide, moderate the session, record and transcribe it, then analyze the results for themes. The method is straightforward; the discipline is in the preparation. Here’s the full process.
1. Define Your Research Objectives
Start by writing down the single decision this research needs to inform. “Should we launch the annual plan at $18 or $24?” is a usable objective; “learn about pricing” is a way to waste ninety minutes. A sharp objective determines who you recruit, what you ask, and, most importantly, how you’ll know whether the session worked. The whole session flows from this one sentence, so don’t skip it to get to the fun part.
2. Recruit and Screen Participants
Recruit participants who match your target audience, and screen them with a short questionnaire to filter out people who don’t fit (or those who do this for a living). A screener should ask a few qualifying questions (relevant behaviors, demographics, product usage) and disqualify professional respondents and anyone with a conflict of interest. Try to ensure everyone in the group is on a similar level so people feel comfortable being honest; a junior employee will not speak freely in front of their VP.
Over-recruit by 10–25% because no-shows are a law of nature, not an exception. Most participants expect some kind of incentive, like cash or a gift card scaled to their time , and specialized or executive audiences cost considerably more to land.
If you’re pulling participants from your own customer base, the same customer feedback workflows you already use can double as a recruiting pool.
3. Write a Discussion Guide
A discussion guide is the moderator’s roadmap: an ordered set of open-ended questions that moves from broad and easy to specific and probing. Open with a low-stakes warm-up to get everyone talking, then funnel toward the questions that actually matter, saving the most important ones for when the group is more comfortable but not yet tired.
Good focus group questions share a few traits, per Krueger and Casey: they’re open-ended (no yes/no answers), they use plain language instead of jargon, and they avoid signaling a “correct” response.
A useful and slightly counterintuitive tip from the same research: go easy on “why” questions, which can feel like an interrogation and push people toward rationalized rather than honest answers. Ask “What went through your mind when…” instead of “Why did you…“. Aim for roughly 10 to 12 questions for a ninety-minute session; more than that and you’re rushing.
4. Moderate the Session
The moderator’s job is to keep the discussion on track, draw out quiet participants, contain dominant ones, and stay strictly neutral. You want to avoid nodding along or leading the discussion. Open by explaining the ground rules, confirming consent to record, and reassuring people there are no wrong answers. Then, mostly, get out of the way. The best moderators talk least; they ask a question, let silence do the work, and resist the urge to fill every pause.
Two classic hazards to manage. The dominant participant who answers everything first: gently redirect with “Thanks. I’d love to hear from someone we haven’t heard from yet.” The groupthink spiral, where one confident opinion becomes a false consensus: probe for dissent directly with “Does anyone see it differently?” A moderator who can tell genuine agreement from polite nodding is gold.
5. Record and Transcribe the Session
Record the session (both audio and video, if possible) because no human can moderate well and capture every quote accurately at the same time. Trying to run the discussion while scribbling notes means you’ll do both jobs badly and miss the exact phrasing that makes a quote usable later. Get explicit, informed consent to record before you begin; if any participant is in the EU, that recording is personal data under GDPR from the moment you hit record, so consent isn’t optional (university research offices like Oxford Brookes publish clear guidance on what that consent needs to contain).
This is where modern tooling earns its keep. Instead of hiring a transcriptionist or transcribing hours of discussion by hand, a tool like tl;dv sits in your Zoom, Google Meet, or Teams call, records it, and produces a time-stamped transcript with automatic speaker detection. When six people are talking over each other, this is a life-saver. Recording and transcription are free and unlimited on tl;dv’s Free plan, so the raw capture costs you nothing but the click.
6. Analyze the Results
Analyze focus group data through thematic analysis: read the transcripts, tag recurring ideas with codes, and group those codes into themes that answer your original objective. You’re hunting for patterns across participants and sessions: the pain point that surfaced in every group, the feature three people independently asked for… You do this while separating genuine group consensus from one loud person’s opinion. Resist the temptation to over-index on the most quotable line; a vivid quote isn’t the same as a representative theme.
This stage is where focus groups traditionally go to die, buried under hours of recordings nobody re-watches. Keeping everything in one searchable research repository is half the battle. The other half is speed: because tl;dv also runs AI over the transcript, you can ask questions across the session — “What did people say about pricing?” — and jump straight to the timestamped moments instead of scrubbing through video. It won’t do your thinking for you, but it turns “a week of analysis” into an afternoon. You can also ask it questions across your entire batch of focus group meetings, making your life ten times easier.
Advantages and Disadvantages of Focus Groups
The main advantage of focus groups is depth. They reveal the reasoning and language behind opinions and surface unexpected insights through group interaction.
The main disadvantage is that they’re not representative. A handful of people can’t stand in for a market, and group dynamics can distort what people actually think.
Both things are true at once, which is why focus groups work best paired with quantitative methods rather than treated as gospel.
| Advantages | Disadvantages |
|---|---|
| Reveals the why behind opinions, not just the what | Results aren't statistically representative of a wider population |
| Group interaction sparks ideas individuals wouldn't raise alone | Vulnerable to groupthink and dominant-participant bias |
| Fast, flexible, and relatively cheap versus large studies | Social desirability bias — people say what sounds good in a group |
| Great for early-stage exploration before you commit resources | Moderator skill heavily influences quality and neutrality |
| Captures real language you can reuse in marketing and product | Analysis is time-consuming and open to interpretation |
At the end of the day, focus groups are a discovery tool. Use them to generate hypotheses and understand your audience’s world, then validate the important findings with a survey or a test. Anyone claiming a focus group “proved” something is misunderstanding what they do.
Focus Group vs. Survey vs. Interview
The difference comes down to depth versus scale and group versus individual. A survey measures how many at scale with quantitative data; a one-on-one interview goes deep with a single person and no group influence; a focus group sits between them, trading statistics for the insights that emerge when people react to each other.
| Focus Group | Survey | Interview | |
|---|---|---|---|
| Data type | Qualitative | Quantitative | Qualitative |
| People involved | 6–10 at once | Dozens to thousands | One at a time |
| Depth of insight | High | Low | Highest |
| Statistically representative | No | Yes | No |
| Group interaction effect | Yes | No | No |
| Relative cost | Medium | Low | Medium–High |
| Best for | Exploring the "why" as a group | Measuring how common something is | Deep individual reasoning |
In practice these methods work best as a sequence: run focus groups or interviews to figure out what to ask, then a survey to find out how many agree. Qualitative and quantitative research work together to answer different halves of the same question.
How Much Does a Focus Group Cost?
A traditional in-person focus group typically runs from a few thousand dollars up to $10,000 or more per group once you add facility rental, professional recruiting, participant incentives, and a moderator. Online focus groups are noticeably cheaper, since you drop venue and travel costs entirely and can often recruit and moderate in-house.
The big variables are recruiting difficulty (niche or executive audiences cost far more), incentives (scaled to time and rarity), and whether you hire a professional moderator or run it yourself.
The costs you can cut without hurting quality are the logistical ones: running remotely removes the facility, and letting a tool handle recording, transcription, and note-taking removes the transcriptionist and the hours you’d otherwise spend on cleanup. A lean, well-run online focus group can cost little more than the incentives themselves.
Time to Run Your Focus Groups
A focus group is one of the sharpest tools you have for understanding why people think what they think. Remember, though, it’s a discovery method, not a verdict. Nail the objective, recruit the right people, moderate with restraint, and pair the findings with quantitative validation, and a single afternoon of discussion can reshape a roadmap.
The part that used to make focus groups a slog was transcription and note-taking. But we’re in 2026. That part has changed the most. tl;dv records and transcribes your session for free, tags who said what, and lets you interrogate the transcript with AI so the insights don’t die in a folder of unwatched recordings. You run the group; tl;dv will handle the paperwork.
FAQs About Focus Groups
How long does a focus group last?
Most focus groups last 60 to 90 minutes. That’s long enough to move through 10–12 questions and let the discussion breathe, but short enough that participants stay engaged before fatigue sets in and answers get thin.
Are focus groups reliable?
Focus groups are reliable for generating qualitative insight and hypotheses, but not for producing statistically representative results. Their findings should be validated with quantitative methods like surveys before you make a high-stakes decision on them alone.
Can AI transcribe and analyze focus groups?
Yes. AI meeting tools can record a focus group, produce a speaker-labeled transcript, summarize key themes, and let you search or ask questions across the session. Combined, it turns hours of raw discussion into searchable, analyzable data that takes a few minutes rather than a few days.
Should focus group participants know each other?
Generally no. Strangers speak more freely and are less likely to fall into established social dynamics. Participants should share relevant characteristics (so they’re comfortable) without being friends, coworkers, or family, which can distort the discussion.
What does a focus group moderator do?
A moderator guides the discussion, keeps it on topic and on time, encourages quiet participants, manages dominant ones, and stays neutral so they don’t lead the group toward a particular answer. Good moderators talk as little as possible.
