What are Focus Groups?

A focus group is a small group discussion, usually somewhere between six and ten people, led by a moderator who asks questions about a product, an ad, a brand, or an idea we’re thinking about launching. The goal isn’t to count how many people like or dislike something. The goal is to hear how people talk about it, what words they use, what confuses them, and what they argue about among themselves.

That last part matters more than it sounds. A survey gives us one person’s answer in isolation. A focus group gives us a room where one person’s comment can trigger another person to disagree, or to say “actually, now that you mention it,” and suddenly we’re hearing a reaction we never would have thought to ask about directly.

Where Did Focus Groups Come From?

The technique has a longer history than most students expect. The sociologist Robert Merton developed what he called the “focused interview” during the Second World War, working at Columbia University’s Bureau of Applied Social Research. He and his colleagues Marjorie Fiske and Patricia Kendall published the method formally in a 1956 book called The Focused Interview. Merton was originally using group discussions to understand how audiences reacted to wartime propaganda films, not to sell soap or soda.

Over the following decades, commercial researchers picked up the technique, adapted it, and gradually started calling it a “focus group” instead. Merton himself later said, somewhat wryly, that he didn’t remember ever using that exact term. So the method marketing students learn about today is really a business-world descendant of a wartime academic tool, which is a fun bit of trivia but also a useful reminder: the method was built to understand how people react and talk, not to produce a clean number.

What Are Focus Groups Actually Good For?

We use focus groups when we want to understand the why behind a reaction, not just the what. If we already know, from sales data or a survey, that a new flavor is underperforming, a focus group can help us figure out whether that’s because of the taste, the packaging, the price, or something else we hadn’t even considered.

They’re also useful early in the process, before we’ve committed to a direction. Say a fast-food chain is thinking about adding a plant-based burger to the menu. Before spending money on a full launch, we might run a few focus groups showing people the concept, maybe even a prototype to taste, and just listen. Do people bring it up unprompted when asked what’s missing from the menu? Do they assume it’s healthier than the beef version (even if it isn’t)? Do vegetarians in the room say they’d actually buy it, or do they say it still doesn’t feel like it’s “for them”? None of that comes through cleanly in a multiple-choice survey.

Focus groups are also good for testing language. Two taglines might test about the same on a survey scale, but in a room we can hear which one people actually repeat back in their own words a few minutes later, which tells us something about what’s sticking.

How Do We Plan a Focus Group?

Running a good focus group starts well before anyone sits down in a room.

Who Do We Recruit?

This is probably the single biggest thing that can go wrong. If we’re testing a product aimed at busy parents, and we accidentally recruit a room full of college students because they were easier to reach, the whole session is close to worthless. We need to define our target segment clearly and screen participants against it: age, income, current brand usage, whether they’ve bought a competitor’s product recently, and so on. A recruiting screener with the wrong questions can quietly wreck months of planning.

We also need to think about group composition. Mixing people who are extremely loyal to a brand with people who’ve never used it can work, but it can also mean the loyal customers dominate the conversation and everyone else just nods along. Sometimes it’s better to run separate groups for different segments and compare what came out of each.

How Many Groups Do We Need?

One group is basically never enough. If we only run one session, we can’t tell whether what we heard reflects the market or just reflects one dominant personality in that particular room who happened to shape the whole discussion. Most researchers run at least three or four groups per segment before they start trusting a pattern, and even then, we’re talking about a few dozen people total, not a representative sample of the market.

Who Runs the Session?

The moderator’s job is harder than it looks. A good moderator asks open questions, follows up on interesting comments, and resists the urge to lead people toward the answer the client is hoping for. A weak moderator can accidentally signal what response is “correct,” and once that happens, the group starts agreeing with each other instead of giving honest reactions. This is why a lot of companies hire an outside research firm to moderate, even when the marketing team could technically run it themselves. An outsider has less riding on any particular answer, and participants sometimes speak more freely to someone who isn’t from the company.

What Happens During the Session?

A typical session runs 60 to 90 minutes, usually with a discussion guide the moderator works through, but with enough flexibility to chase down an unexpected comment. Sessions are recorded, and often observed live from behind a one-way mirror or on a video feed, so the marketing team can watch reactions in real time rather than waiting for a summary report.

One thing worth knowing for an exam or assignment: participants are usually paid an incentive, partly to compensate for their time and partly because paid participants tend to show up and stay engaged. This is a real cost to factor in when we’re pricing out a research project, alongside moderator fees, the facility, and recruiting costs.

What Are the Limits of Focus Groups?

This is where we have to be honest with ourselves about what qualitative research can and can’t tell us.

The biggest limit is that a handful of conversations, even four or five well-run groups, is not a statistically reliable sample. We can’t take a comment that three out of eight people made in a group and conclude that 37.5% of the market feels that way. Findings from focus groups tell us what reactions exist and roughly how common they seem to be, not precise numbers we can build a forecast on.

There’s also a well-known gap between what people say in a group and what they actually do later with their own money. People are naturally a bit performative in a group setting, especially when a topic touches on health, price sensitivity, or anything that feels socially loaded. Someone might say they’d definitely pay extra for the more sustainable packaging, then go buy the cheaper option at the store a week later.

Coca-Cola’s 1985 launch of New Coke is a classic teaching example of this gap, even though the research behind it was mostly blind taste tests rather than focus groups. The company ran an enormous amount of taste testing, and on taste alone, the new formula tested well. What the research didn’t capture was how attached people were to the idea of the original Coke existing at all.

The backlash wasn’t really about flavor. It was about what the company had taken away. It’s a good reminder that the specific question we ask in research matters as much as the method itself; asking “which tastes better” is a different question from “how would you feel if we discontinued the one you grew up on,” and a room full of people talking freely is often better at surfacing that second kind of reaction than a scored taste test.

What Do We Do With the Results?

Once the sessions are done, we’re left with hours of recordings and a moderator’s summary, and the temptation is to just pull out the quotes that support whatever direction we already wanted to go. That’s a real risk, and a good research partner will push back on cherry-picking.

The more useful approach is to look for patterns that show up across multiple groups, not just a single memorable line from one talkative participant. We also want to pair focus group findings with quantitative research where we can, so a concept that tests well emotionally in a group can then be checked against a larger survey to see if the reaction holds up at scale before we commit real budget to it.

From a budget standpoint, focus groups are relatively cheap compared to a full product launch or a national ad campaign, which is exactly why they’re used early, as a way of catching a bad idea, a confusing message, or a name that means something unfortunate in a local market, before it’s expensive to fix.

Key Points to Take Away

1. A focus group is a small, moderated discussion (usually 6 to 10 people) built to explain why people react the way they do, not to produce a statistically valid number.
2. Recruiting the right participants matters more than almost anything else in the process; a poorly screened group can make the whole session useless.
3. Run more than one group per segment. A single session risks reflecting one dominant personality in the room rather than a real pattern.
4. Focus group findings are directional, not precise. Don’t treat a comment from three out of eight participants as a market percentage.
5. What people say in a group and what they later do with their own money can differ, so pairing qualitative findings with quantitative research reduces the risk of acting on talk that doesn’t hold up.
6. Focus groups are usually cheap relative to a full launch, which is why they’re best used early, to catch problems before they become expensive.