Focus Groups: What They Tell You and What They Don’t

A focus group is a moderated discussion with a small group of people, typically six to ten participants, recruited to represent a target audience. A trained facilitator guides the group through questions about a product, service, brand, or concept, and the responses, both what people say and how they say it, are used to generate qualitative insight that surveys and analytics cannot easily surface.

They are one of the oldest tools in market research, and one of the most misused. Done well, a focus group surfaces tensions, language, and mental models that no dashboard will ever show you. Done badly, it gives you a room full of people performing opinions for a stranger behind a one-way mirror.

Where Focus Groups Fit in Market Research

Market research sits on a spectrum from quantitative to qualitative. Surveys, sales data, web analytics, and search volume tell you what is happening and, to some extent, how often. Focus groups, depth interviews, and ethnographic research tell you why. Neither end of the spectrum is complete without the other, and the mistake most marketing teams make is treating one as a substitute for the other.

If you want to understand the full picture of how your market thinks, behaves, and makes decisions, focus groups are one tool inside a broader research programme. The Market Research and Competitive Intelligence hub covers the full range of methods available to marketing teams, from primary research like focus groups through to competitive intelligence tools and behavioural analytics. A focus group in isolation is interesting. A focus group in context is useful.

The classic use cases for focus groups include concept testing before a product launch, message testing before a campaign goes live, brand perception research when you suspect there is a gap between how you see yourself and how customers see you, and customer experience research when you want to understand the emotional texture of a buying experience. That last one is worth pausing on. Understanding the consumer buying process in granular terms is genuinely difficult from data alone, because data tells you what people clicked, not what they were thinking when they clicked it.

How Does a Focus Group Actually Work?

The structure is more deliberate than most people assume. It is not a group chat. A well-run focus group follows a discussion guide, a document prepared in advance by the research team that maps out the topics to be covered, the order in which to cover them, and the specific questions to ask. The guide is not a script. A good moderator treats it as a framework, not a set of handcuffs.

Recruitment is where most focus groups succeed or fail before anyone sits down. Participants are screened against a recruitment brief that defines the target profile: demographics, category usage, purchase behaviour, attitudes. If the recruitment is sloppy, the group will not represent the audience you care about, and the insights will be misleading. I have seen briefs where the client wanted “regular category buyers” but the screener was so loosely written that participants barely used the category at all. The research looked credible on paper and was almost entirely useless in practice.

Sessions typically run between 90 minutes and two hours. They are usually recorded, with participant consent, and observed either through a one-way mirror in a research facility or via a live video feed. Clients often attend as observers, which is valuable, because hearing a real customer describe your product in their own words lands differently than reading a summary in a deck.

After fieldwork, the moderator or a research analyst reviews the recordings, identifies themes, and produces a report. The quality of that report depends heavily on the skill of the analyst. Pattern recognition in qualitative data is not mechanical. It requires judgement, and judgement can be wrong.

What Focus Groups Are Good At

The single most underrated output from a focus group is language. Not what people think, but the words they use to express it. Early in my agency career, I was working on a campaign for a financial services client. The brief was built around language the client used internally, words like “financial resilience” and “wealth protection.” The focus groups came back with something completely different. Participants talked about “not wanting to be caught short” and “having a buffer.” Those phrases ended up in the creative work almost verbatim, and the campaign tested significantly better than anything built on the original language.

Focus groups are also good at surfacing objections. In a survey, people will often select the polite answer. In a group discussion, with the right moderator creating psychological safety, participants will tell you what actually worries them about a product or brand. That information is commercially valuable in a way that approval ratings are not.

Concept testing is another genuine strength. Showing participants a rough creative concept, a product prototype, or a positioning statement and watching how they respond, what confuses them, what excites them, what they ignore entirely, gives creative and strategy teams something concrete to work with. The feedback is not always comfortable, but it is better to hear it in a research facility than after a campaign has launched.

When LEGO went through its near-collapse in the early 2000s and subsequent recovery, the insight that children valued deep play experiences over sheer product variety came partly from qualitative research with kids and parents. BCG’s account of that turnaround is a useful reminder that listening carefully to customers, in structured settings, can surface strategic insight that financial data alone will never reveal.

What Focus Groups Are Not Good At

This is where marketers and clients tend to overreach, so it is worth being direct.

Focus groups cannot tell you what people will do. They can only tell you what people say they think and feel in a specific social setting, on a specific day, in response to specific stimuli. The gap between stated preference and actual behaviour is one of the most documented problems in consumer research, and focus groups are particularly vulnerable to it because the group dynamic adds a social performance layer on top of the individual response.

They cannot produce statistically significant findings. A group of eight people is not a sample. It is eight people. You cannot extrapolate from a focus group to a population with any mathematical confidence, and anyone who presents focus group data as though it were representative is either confused or trying to make a point they cannot otherwise support. I have sat in debrief meetings where a single participant’s comment was being treated as a market signal. That is not research. That is anecdote with a research budget attached.

They are also a poor tool for validating a decision that has already been made. This is the most common misuse I see. A client has committed to a campaign direction internally, there is political pressure behind it, and a focus group is commissioned to “validate” the approach. The groups are designed, consciously or not, to confirm rather than challenge. The output is used to justify the decision rather than inform it. That is expensive theatre, and it wastes everyone’s time.

The research community has written extensively about groupthink and social desirability bias in focus group settings. Participants modify their answers based on what they think the group expects, what they think the moderator wants to hear, or simply what makes them look good in front of strangers. A skilled moderator can reduce this effect, but they cannot eliminate it.

Online Focus Groups: What Changes and What Stays the Same

Remote focus groups, conducted via video conferencing platforms, became standard practice during the pandemic and have remained common because they reduce cost and remove geographic constraints. You can recruit participants from multiple cities, or multiple countries, without paying for travel, facilities, or incentive logistics at scale.

What changes is the social dynamic. The physical cues that a moderator reads in a room, body language, eye contact, who leans forward when a topic is raised, who goes quiet, are harder to read on a grid of video tiles. Participants are also in their own environments, which can be both an advantage and a distraction. Some research suggests that people are slightly more candid online because the screen creates a degree of psychological distance, but the evidence is mixed and depends heavily on the topic and the audience.

Asynchronous online focus groups, where participants respond to questions over a period of days via a text or video platform rather than in a live session, are a different format again. They give participants time to reflect before responding, which can produce more considered answers, but they lose the spontaneous back-and-forth that often generates the most interesting moments in a live group.

The fundamentals do not change regardless of format. Recruitment quality, moderator skill, and the clarity of the research question determine whether you get insight or noise. The platform is secondary.

How Many Groups Do You Actually Need?

The standard answer in qualitative research is that you run groups until you reach saturation, meaning the point at which additional groups are producing no new themes. In practice, most commercial focus group projects run between two and six groups, with the number determined by how many distinct audience segments need to be covered.

If you are testing a single concept with a relatively homogeneous audience, two or three groups may be sufficient to identify the main themes. If you are researching a brand that spans multiple customer segments with meaningfully different attitudes and behaviours, you will need separate groups for each segment, because mixing segments in a single group suppresses the differences you are trying to understand.

Running a single focus group is rarely a good idea unless it is genuinely exploratory and you are treating the output as directional rather than conclusive. A single group is too exposed to the specific dynamics of that particular set of participants on that particular day. One group might happen to include a dominant personality who shapes the entire discussion. A second group gives you a check on that.

Budget is obviously a constraint. A full-service focus group, including recruitment, facility hire, moderation, and analysis, can cost several thousand pounds or dollars per group depending on the audience and location. That is not a trivial commitment for a small marketing team. The question to ask is not whether you can afford focus groups, but whether the decision you are trying to inform is worth that level of investment. If the answer is no, there are cheaper qualitative methods, including depth interviews and online surveys with open-ended questions, that can get you some of the way there.

The Moderator Is the Method

This is the thing most briefs underspecify. The quality of a focus group is determined more by the moderator than by any other single factor, including the discussion guide, the recruitment, or the analysis. A skilled moderator creates conditions where participants feel comfortable saying what they actually think. They probe without leading. They manage dominant personalities without silencing them. They notice when a topic is generating discomfort and decide whether to push into it or let it go. They keep the session on track without making participants feel managed.

That is a genuinely difficult skill set, and it is not the same as being a good presenter or a good interviewer. I have seen senior marketing people offer to moderate their own focus groups to save money, and the results are almost always compromised. When a client moderates their own research, participants can usually sense the relationship, even if it is not stated, and they adjust their answers accordingly. The independence of the moderator is not a formality. It is a structural requirement for the method to work.

When briefing a research agency or independent moderator, the most important thing you can do is be honest about what you do not know and what you are genuinely uncertain about. A moderator who understands the real question, not the sanitised version of it, will design a better discussion guide and probe more effectively in the room. Vague briefs produce vague research. That is as true for focus groups as it is for creative briefs or media briefs.

The relationship between brief quality and output quality is something I feel strongly about across all of marketing. I have written before about how much strategic waste happens upstream of execution, in the brief itself, before a single pound of budget is spent. Focus groups are no different. A weak brief going into a focus group will produce findings that feel interesting but answer the wrong question.

How to Use Focus Group Findings Without Overreaching

The most disciplined thing you can do with qualitative research is to be explicit about what it can and cannot tell you, before you share the findings with anyone who might use them to make a decision.

Focus group findings should be framed as hypotheses, not conclusions. They should prompt questions, not close them. If a focus group suggests that customers are confused by your pricing structure, that is a hypothesis worth testing with a larger quantitative sample, not a finding that justifies a pricing overhaul on its own. If a group responds enthusiastically to a creative concept, that is a signal worth noting, not a green light to produce and run the campaign without further testing.

The most productive use of focus group output is to feed it forward into the next stage of research or into a creative or strategic process. The language participants use becomes copy testing material. The objections they raise become survey questions. The confusion they express becomes a UX brief. The findings are most valuable when they are treated as inputs to something, not outputs from something.

One practical discipline I have found useful is separating the debrief into two parts: what we heard, and what we think it means. The first part should be as close to verbatim as possible. The second part is interpretation, and interpretation is where bias enters. Keeping those two things separate in the room, and in the report, makes it easier to challenge the interpretation without dismissing the evidence.

If you are building out a broader research capability or thinking about how qualitative methods like focus groups connect to your competitive intelligence work, the Market Research and Competitive Intel hub covers how these methods fit together in practice, including where quantitative tools pick up where qualitative ones leave off.

When to Choose a Different Method Instead

Focus groups are not always the right tool, and part of being a competent research buyer is knowing when to choose something else.

If the topic is sensitive, personal, or socially charged, individual depth interviews will almost always produce better data. People will not talk honestly about their debt, their health anxieties, or their relationship with alcohol in a room full of strangers. The group format actively suppresses disclosure on sensitive topics.

If you need statistical confidence, you need a survey, not a focus group. If you need to understand actual behaviour rather than stated attitudes, you need behavioural data, observational research, or a well-designed experiment. If you need to understand how people interact with a digital product, usability testing will give you more actionable insight than a group discussion about it.

The honest question to ask before commissioning a focus group is: what decision will this research inform, and is a focus group the best way to get the information that decision requires? If the answer to the second part of that question is yes, proceed. If the answer is “we always do focus groups at this stage,” that is a process, not a rationale.

Marketing teams that treat research as a ritual rather than a tool end up spending budget on findings they already expected and using them to confirm decisions they had already made. That is not research. It is reassurance with a fieldwork invoice. The value of a focus group, like the value of any research method, is proportional to how genuinely open you are to being surprised by what you find. If you already know what you want the research to say, you do not need the research.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions