Discovery Methods
: CHAPTER #
6

Focus Groups

This chapter explores what focus groups are, what their role in UX research is, and how to conduct focus group studies.

A focus group is just what it sounds like: a discussion group convened and conducted with a specific focus in mind. It could be plumbers discussing the installation of shower heads, or frozen treat lovers discussing their preferences in the texture of yogurt, or pet owners discussing their grievances with fur around the house, or cancer patients sharing their insights into what it’s like to go through treatment at a certain stage. You get the idea.

A focus group is not an interview that happens to have multiple respondents in the same room. Au contraire, the very value of a focus group resides in the group discussion, and the interactions that transpire among the members.

A focus group is best used to explore what people believe, how they feel, and what they perceive. Other research methods, like surveys for example, are best applied when you’re interested in what people do and why.

When to choose focus groups

Focus groups have a very specific role in the research realm, because they provide some kinds of data and not others; a focus group can tell you what some people are thinking, but not how many people are thinking it. For example, a focus group can tell you how some people go through a complex process, such as deciding which product to buy, or why the "help" feature of your product isn't helpful, or what messages your branding efforts actually get across.

If you need to find out whether the responses you get are typical—does everybody hate the help desk feature?—you can follow up with some form of quantitative study. If not, you can use the qualitative results of the focus group on their own, as seeds for further development.

Focus groups are great when you don’t know a whole lot about how your product or service is perceived, and you’d like to learn more; when group dynamics are the best way to get at answers because banter will reveal qualities that a one-on-one conversation might not; and when you’d like to dig in on multiple topics in one research cycle.

When in the product development cycle to use focus groups

Focus groups are great to pair with other research methods at all stages of development, but especially so in discovery.

For example you might begin with some generative interviews to ideate, and once you have a loose idea of the product you wish to build, host a focus group to further explore and define the problem you wish to solve with your product-build.

You can also employ focus groups to test products already in development—the product in question need not be finished. You can run just certain aspects through focus groups to see if things are working independently.

You can also use focus groups to explore marketing and branding. Groups can brainstorm ideas, images, and associations at the beginning of the branding process, and they can also be used to test whether a given campaign actually conveys the message that it's supposed to.

Benefits of focus groups in UX design

Focus groups are a good way to draw out ideas and information that participants might not be willing or able to share in an interview setting or through a survey. 

Being surrounded by a small group of peers makes many people more comfortable and more willing to talk, even about sensitive or personal topics. 

The interaction among group members can bring out ideas that were previously not conscious, or simply taken for granted. The way group members agree or disagree—either with passion or indifference—can shed light on how important a certain consideration should or should not be to your development team.

Potential pitfalls of focus groups

Focus groups cannot provide quantitative data, like whether a certain viewpoint is typical of a given category of people—American plumbers, British beekeepers, female homemakers in Georgia, six-year-old boys, etc. Like other qualitative research, they lack statistical specificity and accuracy. Sorting and parsing the feedback and banter of a group of people can be time consuming and complicated, and your own biases can get in the way as you’re sorting data (like whether you disliked a certain participant, for example).

The success of a focus group also depends heavily on the skill of the moderator.

There are other limitations and circumstances to be aware of. They are manageable pitfalls, but if left unmanaged, the below examples describe circumstances when focus groups either can't be convened or would not give good results.

1. When there is no practical way to gather enough participants

If the people you want to talk to are scattered thinly all across the country, it may be impossible to gather enough of them together in one place to have a focus group. While teleconferencing might be an option in some cases, generally the awkwardness of such technology would prevent any real group cohesion.

2. There are cultural barriers to open discussion

While Americans are often very comfortable saying what they think, even if their thoughts are critical, some cultures have no tradition of airing criticism, especially in the presence of a perceived authority figure like a group moderator. Where cultural mores may clash with being honest, a focus group will not give you any usable data.

3. There are cultural barriers to discussion of the topic

Even a culture happy to talk about some things has topics considered unmentionable. A product that might be deemed “embarrassing” for one reason or another might not be a good candidate for discussion by a focus group, unless you’re expressly trying to explore those specific dynamics.

4. The participants are competitors

Avoid defining your groups in such a way that business people find themselves in the same group as their competition, especially not if you want to ask about potentially sensitive or proprietary topics. For this reason it’s generally good to screen competitive or industry professionals from your participant base.

5. The prospective participants are not equals

There are things people won't say in front of their supervisors, teachers, clients, or others in a position of power. Seniority in an industry, or differences of race, gender, ability level, or socioeconomic class can also create power differences that could alter group dynamics. It's not that any power imbalance at all precludes the usefulness of a focus group, but avoid extreme imbalances. Also avoid imbalances that are likely to interact badly with the discussion topic and situations that are likely to exacerbate an imbalance, such as a group where every member except one belongs to the same privileged category.

In some of the above cases, a skilled moderator might be able to make a challenging group work anyway. There are no absolute rules. The point is that the main drawback of focus groups is that there will be times you want to use them and find that you can't.

And finally...

6. Group dynamics are tricky

You might have screened and selected the most perfect possible participants, and still, regardless of your perfect candidates, people can sometimes get lost in an idea. Some tend to jump onto ideas they don’t even believe in, or become devil’s advocate and argue against an idea they really like. Shy people are less heard, compelling people can be persuasive even when they’re wrong, and sometimes, like a bad dinner party, there’s just no group chemistry no matter how hard your moderator might try. In most of these cases, a strong moderator will help. In others, cut your losses and move on. If a group isn’t working, it’s not working. End the group if it’s not useful and try again another day.

For these reasons, you should not put the weight of all your research on focus groups alone. Information gleaned from a focus group should not on its own dictate significant courses of action or significant budget allocations.

Planning

There are a lot of decisions to make before you hold focus group sessions, beginning with what the group will be about and what kind of people you need to invite.

  • Do you need to talk to people in a given industry (plumbers? teachers?) or in a particular socioeconomic group?
  • Does education level matter?
  • Does gender matter?
  • Who has the information you need?

Once you have the basics down, you can tackle more detailed questions, like how many people belong in each group, how diverse each group is, and how many groups you want to conduct.

How many people do you need per group?

The important thing about group size is how it affects the group dynamic. Too few people, and the discussion won't take off. The meeting will function more like an interview with multiple interviewees. Too many people, and not everyone will be heard.

Rule of thumb: Three and 12 are generally considered the outside limits, but opinion varies greatly on where, exactly, the sweet spot in the middle lies. Since some people usually fail to show up (roughly 10-15%), it's usually better to invite too many than too few, as if too few show up, you'll have to cancel.

What kinds of people should be in each group?

If you have a sense of your product’s target personas, there may be an opportunity to match your participants with those personalities. If you’re in earlier stages of development, a diverse population within your product’s target market will be best.

The diversity of participants will depend entirely on your target market. All sorts of people wear sneakers, for example, so a focus group on sneakers would be very diverse indeed: a whole spectrum of sneaker-wearers could be included. A focus group on tie clips and cufflinks, on the other hand, might look a bit more homogenous. If your target audience is tighter, you might need fewer groups overall.

How many groups do you need?

Rule of thumb: 4-6 groups is the average, though some opt for smaller studies and do just 2-3, and some opt for more and can do as many as 15.

The more groups you have, the more ideas and opinions you will collect, but this is helpful only up to a point. Go beyond six, and additional groups are likely to tell you more of what you already know.

How many groups your study will need depends on your budget, how many different varieties of participant you want to include, and many different geographical areas you want to cover.

How to organize your thoughts around focus group planning

Starting from averages, figure you want 8-10 people in each group, and that you’ll hold 4-6 groups overall. Then start parsing potential participants using the criteria described above, and if you find you need more people or fewer groups or both or vice versa, adjust from there.

Consider how to blend and divide your participants. Disagreement can be useful, so make each group as diverse as possible (again, within the target audience of your product).

Be prepared to host rounds of focus groups. For example, if you think men and women will use your product differently, you can consider 2-5 focus groups of women and 2-5 focus groups of men.

Where is the group going to meet?

Location is partially a question of logistics; you need some place everyone can get to, and you need a comfortable room of the right size that is free of distracting noise.

The group dynamic depends largely on a space where participants feel comfortable, and different people feel comfortable in different kinds of places. For example, businesspeople will find the conference room at a hotel comfortingly familiar, whereas another group of participants might prefer a ring of folding chairs in a space that feels like a church basement or a classroom. The question is: Under what circumstances do these people normally sit around and talk about this subject? Then mimic those circumstances. Nobody likes to feel like fish out of water.

If the people you want to recruit might be attending a conference or convention together, it could be especially convenient to recruit and hold your study at the event, on premises. If you’d like to observe from behind a two-way mirror, there are services in most cities that can provide these sorts of spaces.

When should the group meet and for how long?

A group should meet for about an hour to an hour and a half. However, your participants will also have to take time out of their day to travel to and from the meeting site, so their total time commitment is likely to be two hours or more. What time of day is best for dropping everything for two hours? That is your critical question, and the answer will vary from group to group. If you want to talk to stay-at-home parents, schedule the group while the kids are at school. If you want to talk to many working people, you will have to pick a time outside of business hours. Although, if your participants have the flexibility to take time off in order to attend the group, a daytime meeting will feel familiar and sensible.

Remember set-up and clean-up time when you decide how long to book the venue for.

Who will moderate?

Assuming you have multiple properly trained people to choose from, who should moderate depends on the group. In most cases, the moderator should seem like, if not actually be, a member of the category of people you want to talk with.

For example, if your participants are likely to show up wearing suits, the moderator should wear a suit. If the participants are more likely to wear jeans, so should the moderator. The moderator need not wear a disguise and go undercover, but minimize opportunities for participants to perceive him or her as "one of Them."

Although the role of moderator does not require any special expertise in the subject being discussed, in some cases insider knowledge is an important part of the moderator being accepted by the group.

A moderator can make or break your study. If they’re too passive, the group can get lost in its own momentum and veer off topic. Too leading, and your results will be skewed. Awkward, and the participants won’t feel comfortable. Overly confident, and they could be perceived as domineering and off putting. So interview your moderator first, and get clear with them on what you want and get a sense of how they operate.

Logistical preparation

Besides your initial recruitment efforts and booking the venue, there are many small tasks that must be completed to make the focus group work well.

  • Visit the location. Is it everything it is supposed to be?
  • Is the room easy to find? If not, find or create a map.
  • Is the building easy to find?
  • Is there affordable parking nearby?
  • Does everyone on your team know who is doing what and when?
  • Is anything problematic likely to be happening in or near the venue during your group meeting? For example, is an unrelated convention likely to make the place confusing, crowded, or noisy? Is a sports game or college commencement nearby likely to snarl up traffic?

Send at least two email reminders, one a week or so ahead and one a day or two before, containing all relevant details (including directions) to everyone who has agreed to participate. The day before the study, make a reminder phone-call or send a reminder text, and offer to answer any remaining questions. Include your number in any emails so people can get in touch with you if they need to. Many services will handle this for you, but you might want to make sure this is part of what they offer.

Conducting a focus group

The moderator's job is not to ask questions and get answers the way an interviewer does. Instead, the moderator uses questions, group exercises, and other techniques to get the discussion going and keep it going. Where an interviewer may work  from a script, the moderator works from a topic guide and steers the conversation through the various subjects the group has been convened to discuss.

What questions the moderator asks may be indirect. For example, say you want to know why participants make certain choices. The difficulty is that most people don't know why they do things, and asking why is likely to prompt them to make something up that sounds good. For example, a person might actually prefer a given brand because the local store tends to stock it right next to the cookies. Anything associated with cookies has to be good, right? But pressed for an explanation, the person is likely to talk intelligently about performance and cost. An indirect question, like "walk me through a typical trip to the store" has a much greater chance of getting at the truth.

Perhaps the trickiest part of the moderator's art is to make sure that everyone gets to speak and nobody monopolizes the conversation. It is not enough to know who isn't speaking very much, the trick is to figure out who would be speaking more if only they felt more comfortable speaking, or if only the pace of the conversation were a little slower. For example, if someone brings up an idea and everyone else ignores it, the moderator has to know whether some kind of power dynamic is at play, and what to do about it without making everyone the group too uncomfortable to participate.

The conversation should be both audio and video recorded, since group discussions are too complex for effective note-taking, and the participants must be reminded at the beginning that they are being recorded.

Analysis

Analyzing group discussion information can be difficult, in part because there is so much of it. Transcripts alone can run to dozens of pages, too much for casual study to make sense of without introducing some kind of bias or leaving something important out.

Fortunately, there are many forms of appropriate qualitative analysis to choose from. Generally, the analysis process must include both the video and the transcript, in order to capture non-verbal communication and in order to eliminate confusion over who said what.

Additionally, focus groups tend to reveal big-picture information (as opposed to, say, the minutia of what you’d observe in a field study) which makes things easier. Chances are, you went into your focus groups wanting answers to specific questions or insights on specific topics. Look for these as the conversation unfolds and compare what one participant said to another, or how people agreed and disagreed with one another, or for more subtle cues around your topics of interest, like ideas that speak to the issues you’re interested in without naming them specifically. Sometimes conversations on one topic can point to solutions on another.

Conducting a focus group

The moderator's job is not to ask questions and get answers the way an interviewer does. Instead, the moderator uses questions, group exercises, and other techniques to get the discussion going and keep it going. Where an interviewer may work  from a script, the moderator works from a topic guide and steers the conversation through the various subjects the group has been convened to discuss.

What questions the moderator asks may be indirect. For example, say you want to know why participants make certain choices. The difficulty is that most people don't know why they do things, and asking why is likely to prompt them to make something up that sounds good. For example, a person might actually prefer a given brand because the local store tends to stock it right next to the cookies. Anything associated with cookies has to be good, right? But pressed for an explanation, the person is likely to talk intelligently about performance and cost. An indirect question, like "walk me through a typical trip to the store" has a much greater chance of getting at the truth.

Perhaps the trickiest part of the moderator's art is to make sure that everyone gets to speak and nobody monopolizes the conversation. It is not enough to know who isn't speaking very much, the trick is to figure out who would be speaking more if only they felt more comfortable speaking, or if only the pace of the conversation were a little slower. For example, if someone brings up an idea and everyone else ignores it, the moderator has to know whether some kind of power dynamic is at play, and what to do about it without making everyone the group too uncomfortable to participate.

The conversation should be both audio and video recorded, since group discussions are too complex for effective note-taking, and the participants must be reminded at the beginning that they are being recorded.

Techniques

In some groups, the participants might also watch a video, try out a product or a part of a product, or examine photographs or graphs. Group activities could involve drawing on an easel or a whiteboard, or writing lists. All such physical aides also count as tools.

The common group activities for focus groups include:

  • Brainstorming ideas or word associations
  • Completing sentences
  • Developing a hypothetical marketing campaign
  • Drawing pictures that symbolize various aspects of the topic
  • Answering imaginative questions, such as "if you could change anything about your job, what would you change?" or "if our product were a person, what would he or she be like?"

Focus group hybrid methods

Focus groups are rarely used alone, even if the goals of the research are entirely qualitative. Interviews, for example, are another qualitative research approach that can be used to drill into some of the more micro details (the “whys”) that focus groups aren’t designed to answer.   

Alternatively, a focus group study can be used the develop the parameters of a quantitative study, such as a survey, or a focus group can follow up on a survey.

Tools and analysis

Focus groups require recording equipment, first and foremost. You will definitely want to review footage later.

Other logistical considerations will include questions and requests made by the moderator, the appearance and behavior of the moderator, and even the size and shape of the room, as all of these factors stand to influence group behavior.

Analysis

Analyzing group discussion information can be difficult, in part because there is so much of it. Transcripts alone can run to dozens of pages, too much for casual study to make sense of without introducing some kind of bias or leaving something important out.

Fortunately, there are many forms of appropriate qualitative analysis to choose from. Generally, the analysis process includes both the video and the transcript, in order to capture non-verbal communication and in order to eliminate confusion over who said what.

Additionally, focus groups tend to reveal big-picture information (as opposed to, say, the minutia of what you’d observe in a field study) which makes things easier. Chances are, you went into your focus groups wanting answers to specific questions or insights on specific topics. Look for these as the conversation unfolds and compare what one participant said to another, or how people agreed and disagreed with one another, or for more subtle cues around your topics of interest, like ideas that speak to the issues you’re interested in without naming them specifically. Sometimes conversations on one topic can point to solutions on another.

Social interaction is at the heart of focus groups

In a focus group, your biggest gems are the ways in which people interact with one another. Their purpose is to observe social interaction. They’re powerful ways to get insights into perception and opinions, not just around your topic specifically, but to get a sense of your vertical, your market, and your competition. When they’re the right research method for the task at hand, they can fundamentally alter the direction of a project or product for the better. 

NEXT CHAPTER

Qualitative Usability Testing

This chapter will walk you through the process of conducting qualitative usability tests.
Go to next chapter →

Get the UX Research Field Guide delivered weekly.

Subscribe