Focus Groups Reconsidered

Focus groups have a poor reputation in the design community. They’ve been around since the 1940s and look increasingly old-fashioned against the prevailing wave of ethnographic techniques now in vogue. If you sort through a pack of IDEO method cards you’ll find focus groups omitted with prejudice.

The arguments against focus groups are well known, and my instructors at Carnegie Mellon railed against the technique, but I’ve never really spent much time thinking about the merits. I’m curious because there are some significant parallels with the group-based techniques common to service design and I’d like a better understanding of those dynamics.

The best book I’ve found on the subject is a 1997 publication by David Morgan called Focus Groups as Qualitative Research, part of a fantastic series on research methods. For me, its most thought provoking insight is the notion that focus groups can be used in conjunction with other qualitative techniques.

The author regards this as conventional wisdom and cautions against the “unwarranted assumption that focus groups must be limited to preliminary or exploratory uses in combination with other methods.” And later, “the reputation that focus groups have in some circles as a ‘quick and cheap’ technique is due to the very limited function to which they have often been relegated as preliminary explorations to set the stage for real research.”

To be honest it had never occurred to me that focus groups could be used as anything other than a stand-alone tool. I’ve seen them used in parallel with other qualitative techniques, but not in any sort of integrated fashion. Viewed in isolation the criticisms of focus groups might be compelling (though I think the author would disagree). But as a preparatory step I can see quite a bit of resonance for design researchers.

Focus Groups and Participant Observation

The principle benefit that focus groups have to offer to a project based on participant observation is a concentrated insight into participants’ thinking on a topic. This can be especially useful when the researcher is entering a field site that differs sharply from his or her prior experience.

In this case, the focus groups provide an initial exposure to the typical experiences and perspectives of those the research is about to observe. Given the well known problems of gaining access to and establishing rapport in a new field site, preliminary focus groups with participants drawn from similar locations, other than the research setting itself, can often be quite useful. [Morgan p33]

Focus Groups and Individual Interviewing

The basic idea is to use one or two exploratory focus groups to reveal the range of thoughts and experiences prior to the first individual interview. Alternatively, preliminary individual interviews can help generate focus group discussion guides by giving a feel for how people think and talk about the topics that the groups will discuss.

A different “supplementary” use for either type of interview would be to learn about differences among potential interviewees. Thus, in an individual-interview project that involved a choice among several sites or population groups, a preliminary round of focus groups would provide a basis for selecting the next set of interviews.

A final way to combine focus groups with individual interviews is to conduct one as a follow-up to the other. Following individual interviews with focus groups allows the researcher to explore issues that came up only during the analysis of the interviews. Alternately, followup individual interviews can help provide depth and detail on topics that were only broadly discussed in group interviews. [Morgan p34]

The basic argument in favor of focus groups is that they reveal aspects of experiences and perspectives that would be less accessible without group interaction. Unfortunately this interaction can cause problems of its own and much of the criticism leveled at focus groups centers on this issue.

In any event, the main advantage for designers is the opportunity to observe a large amount of interaction on a particular topic in a limited amount of time. I’m intrigued by the idea that focus groups might also provide a preliminary understanding of an otherwise foreign subject and help set the stage for more detailed research. But it’s clear that the author views this as a limited application of an otherwise robust strategy. “For many purposes, focus groups, like other qualitative methods, can be a well-chosen, self-contained means for collecting research data. … [and] when pursued in this way, focus groups demand the same attention to detail as any other means of data collection. The quality of data depends on the quality of preparation.”

He’s talking about focus groups as a rigorous tool of social science, which I gather is about as far removed from the marketing-based focus groups of scorn as actual ethnographic research is from the watered down version of “design ethnography” that designers know and love.

Throughout the book the author highlights best practices for successful focus groups. I’ve compiled a list of about a dozen practical guidelines from the book for future reference, including a set of techniques for group self-management.

  1. Hi Jeff

    I’m a service designer with a design research background. I can endorse the use of focus groups, with some additions to the solid comments above.

    I think the crux is ‘design of research’. Like any design challenge, you need to design your research to get the output you need. To this extend I’m against treating ‘research methods’ as givens, which many designers seem to do.

    Good research design involves understanding the strengths and weaknesses of the tool. No point in criticising discussion groups when applied to the wrong problem. As an aside, it seems to me the dominance of observation/ ethnography carries the same risk – good tool applied too often, and so often to the wrong problem.

    Blunts the tool and confuses the problem.

    So my question is… ‘how creative can designers be in designing groups (and other research tools) that deliver exactly what they need?’

    Any group discussion is first about what people say, and second about how they relate to others in the group (including the researcher). If you’re into semantics or discourse analysis in any way, discussion groups are great for understanding dominant discourses very quickly. Discourses tell you what’s socially powerful and how it works.

    For me, the point of a group is to create ways to help (potential) users articulate their experiences and desires in useful ways. Groups are of limited use in ‘testing’ established ideas, and of arguable use for understanding behaviours and/ or contexts.

    For example, I use groups alongside other methods to…

    – Begin collaborative work with users, especially early in a design project – perhaps to set up partnerships with a small group of users etc.
    – Sketch key areas of learning, such as user journeys, to scope the problems and opportunities.
    – Model social interactions, such as different types of users in a market and how they interact (this can mean groups of up to 40 people – very carefully selected and managed etc).
    – Explore issues with my business clients – asking users and the client to workshop a challenge together.
    – Explore brand and communications – one area where groups are actually a solid tool.

    In other words, it’s not the groups that are really the problem here. It’s our willingness to treat groups as a given, a method, and ignore the fact that, like everything else, it’s a design artifact.

    Let’s make the group a strategic, adjustable, well-designed tool.

  2. Jeff

    Hi Stephen,

    Your comments about the “design of research” really resonate with me. I wish that more designers understood that aspect of the process. Too often I see design research executed by rote.

    Thanks for your insight into the flexibility of groups methods. I’m always interested in finding new ways to apply these techniques.

  3. Hi Jeff,
    Focus Groups is a qualitative research method i extensively use during the analysis/user research phase of intranet projects i follow.

    Since often happens to me to offer consultancy to company having a core business quite foreign to me and needing an overview on their main processes and subjects, I found Focus Groups a great tool to speed-up my learning of their domain of work.

    Setting up Focus Group for an intranet is quite easy, since the only logistic stuff you have to manage is getting employees (for sure the right ones) in a office meeting room.

    Moreover at the end of a first round-up of Focus Group sessions you get also good insight about which people from which business area will be the best to single interview in the next qualitative user research phase.

    One great side effect about this process in the intranet projects is that by meeting many people from all the company you produce a great widespread engagement.
    This engagement is essential to get the intranet on the right track: those people are both the final users and the producer of the intranet.

    This approach is called by me the “involving approach” and is briefly described by this my case history on the intranet of Honda Italia (is the local company of the famous japanese motor company).

    Finally i found userful to do Focus Group (always in intranet projects) when you need to better understand a conflict: eg. two departments of the company are not able to collaborate. If you make a Focus Group with some people from both and you are able to adress the conversation not directly to the conflict, but on the point where the business process involve both you’ll be surprised about how often the reason of the conflict emerge and above all how often a solution is found there, in that moment.

    I think this happen because a third neutral party (you) is involved and the conflicting parties get for the first time the chance to explain what is not working differently, to let understand to an external player what is happening. Instead usually when they are one against the others they are not able to anlayze differently and less emotionally the conflict, and they just reproduce it in the same way.

  4. “Let’s make the group a strategic, adjustable, well-designed tool.”

    Fantastic comment by Stephen, and great post Jeff.

    “The basic idea is to use one or two exploratory focus groups to reveal the range of thoughts and experiences prior to the first individual interview.”

    This resonates with me greatly. Recently I conducted a focus group which tested the perceptions and capabilities of people to think outside the box and generate new ideas. Preparing for workshops to co-create new ideas requires a huge amount of design in the tasks, generative tools, environment (not to mention the facilitator role which in my opinion is being severely overlooked in design education..and one to importantly consider in the focus groups context) and we have used the focus group to generate insights to design the workshops correctly for the type of people who will be attending.

    In this comment I see similar parallels in what you say Jeff, just in the context of using focus groups to inform the design of a co-creative workshop.

    Would be interested to know your thoughts..

  5. Jeff

    Cristiano, thanks for that example. It’s nice to see some confirmation of this thesis. I’m getting the impression that focus groups might be more common overseas than they are in the US design community.

    And Sarah, it’s good to hear that focus groups can integrate with a service design process. The approach I’m describing is about building context and it sounds like that’s applicable to a whole range of different methods. It seems like anything that takes significantly more time and effort than a focus group could benefit from some preliminary reconnaissance. I do a lot of one-on-one interviews and there’s always a learning curve when I’m entering a new domain. If I write off a handful of interviews the rest usually make up for it. But with a co-design workshop it’s a one-shot method. Pass/fail. Makes sense to get a jump on it if you can.

  6. Angela Meyer


    This is an interesting post and thanks for taking the time to compile the practical guidelines. I do think that focus groups can be a useful tool for designers, but what tends to happen a lot of the time is that this type of research is often hijacked by the client’s marketing group, usually because the client will make the case that their marketing group already has experience with the method and has an existing relationship with a vendor to provide the research.

    In my experience, the critical element for finding value in the focus group method is for the design team itself to run the focus group. I have used focus groups as a research method, and it has proven to be a worthwhile and valid way of getting good intelligence about users, their thinking, and their behaviors. Not to mention that it can be more efficient and economical than visiting users individually. I consider it to be a good “quick and dirty” method. As Stephen pointed out above, focus groups can also be an excellent formative tool for designing research itself.

    However, I have also worked on projects where my client insisted on outsourcing the focus group work, and that’s where you run into trouble. My experience is that even with professional market research firms, the moderator and probing can often be weak, not to mention that selection of the participants needs to be well-scrutinized.

    Usually this is the only research method corporate organizations are familiar with, and it tends to get driven by previously established norms around marketing-related inquiry, rather than design-related inquiry. So it is harder to win this territory back over to the design space once you start going down that track. Maybe the trick is to use the method, but call it something else.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s