Doug Garnett’s Blog


Defend Focus Groups: A Critical Research Tool

<strong>Defend Focus Groups:</strong> A Critical Research Tool

I was part of a brawl on Twitter the other night. You know the kind — over the details and relative value of different market research methodologies.

The next Wednesday morning people got some great laughs from Mark Ritson’s blow by blow police blotter summary of the brawl in MarketingWeek. 

Of course, I’ve never physically brawled. Still, I was one of the primary participants (along with Byron Sharp and Everard Hunder). All three of us (Byron, Everard, and myself) know these issues matter and are important enough to argue about. So, I respect both men quite highly.

Weeks later I remain surprised by how many otherwise good thinkers embrace an anti-focus group attitude. (To be clear, Everard DOES rely on focus groups regularly as part of his company’s research efforts.)

Research Method Must Match the Purpose of the Research.

It was odd to argue solely about a method. Method is not more important than purpose. So, it is irresponsible to discuss “best” method without also asking “for what?” Even methods I have grave concerns about (like dial group research) have some valid applications. As a result, it’s my policy not to say “method X is invalid.”

We need extensive, continuing discussions about the relative value and risks of method X vs method Y in order to do Z. We do not need discussions that start with “method X is invalid”.

Why Do We Even Need Qualitative Research?

Before we can look specifically at focus groups, we need to fundamentally talk about qualitative research. After all, many companies simply don’t value qualitative research.

Many critical product, management and marketing issues, though, can’t be quantified at reasonable cost, within a reasonable amount of time, or ever – regardless of cost and time. In fact, W. Edwards Deming observed:

“the most important figures that one needs for management are unknown or unknowable.” 

Qualitative research work helps us dig into, and around, these many issues.

What are the Big Things which can be Achieved with Focus Groups?

No single type of qualitative research covers all situations — whether focus groups, one-on-one interviews, ethnography, marketing anthropology, or any of the even newer fads. 

But rather than offer some complicated list of “use focus groups for X but not Y”, let me discuss what Carla Roberts (Team 360 Consulting) and I have been able to achieve with groups. This is important because we’ve created a very unique means of leveraging the opportunity groups offer. 

Carla and I first worked together in 1993. I was referred to Carla for her to moderate focus groups for an ad agency where I was head of strategy — this agency specialized in direct response television. Our first project was for Sears where we ended up developing an approach to learning via groups which turned into our own unique research. Critically, it enables us to build a reliable product or message strategy from just one, well conducted, research project (of multiple groups, of course).

It’s our experience that most people (including most researchers) don’t have a clue how much important information they can glean from groups or how to get to it. These low expectations and inadequate methods lead to poorly executed groups that end up not delivering what they should have. 

Our approach to groups is vastly different from the typical research found in most corporations. First, we do something that others do not do — we spend a tremendous time up front creating a strategy for group execution: recruiting, discussion approach, and a stimulus strategy for discovering what we need to know. Just turning people lose to “talk” is an incredible waste of research dollars and rarely uncovers what’s truly ground breaking and important.

Our stimulus strategy is designed around the truth that consumers interact well in groups by responding to ideas. Rather than bring in a copywriter, we create stimuli to approach the ideas along with a flow in which we’re going to wander through the topics. However, to encourage free discussion and direction, we remain flexible and move as needed to whatever arises, anticipated or not. (Stimuli we use range from written material to video, sound, and even physical props.)

Using those stimuli, group participants respond to issues much more clearly than they would without it and are able to dig beyond what we’ve already put in front of them to uncover things we didn’t expect. In the end, the responses to these stimuli lead to conclusions which are far more powerful than we could have discovered in any other way.

We’ve also found that when clients follow along with us, we end up with a team who can execute with clear vision and strong unanimity. This work builds a shared understanding of the insight needed for success (very unlike the extended arguments and bickering we so commonly hear after other research.) 

The Arguments I Hear Against Focus Group

Never thought I’d have to argue for focus groups merely having value — but, apparently, that is where we’ve arrived.

Bruce Clark (marketing prof at Northeastern) was quite helpful when, after the brawl, I was searching to find where anti-group ideas were coming from. He pointed me to a Griffin and Hauser 1993 article published with the imprimatur of “MIT”. This article seems to have outsized influence given its similarity with arguments I heard the other night. (He pointed to a few more articles, one is discussed below.)

The Griffin and Hauser article is exceptionally narrow. It considers focus groups vs individual interviews ONLY for the purpose of generating long lists of ideas from customers for improving products (the length of the list — not its quality — is the key purpose of their work). Griffin and Hauser were pursuing this goal because TQM said it was important.

Is this really the best MIT has to offer? Focus groups are a very bad approach for just generating lists. Even worse, this use of research is a bad idea. To paraphrase Steve Jobs: if we ask customers to tell us what we should do we’ve already lost the battle.

Two claims in the article were brought into the brawl:

Claim 1: Individual interviews are more cost effective (all they found was interviews generated longer lists of ideas per dollar). Ok. I’m not the least bit surprised at the cost result. What I’m surprised by is using any research to merely build lists.

Claim 2: Focus groups only provide a few minutes for each participant to talk – and that’s bad (apparently). Uh. Err. I thought we wanted to learn important things from research. Guess I’ve been working too hard — my career would have been SOOOO much easier buying minutes with customers in the Costco size. Sigh. Two more thoughts:

    • The QUALITY of the minutes is of ultimate importance and very few studies attempt to look at this issue because it’s exceptionally hard to quantify the quality of ideas learned or the value of the insights we take away.
    • Quality is also critical in ethnographic work. For example, I find 90% of the minutes spent in ethnological research are “wasted” according to these terms. They aren’t “wasted” when you realize that the 90% is required to get a valuable 10%.

What I find striking in the various research looking at focus groups is that each study has absurdly narrow goals. Narrow goals are the only way to get an article published, though. This also indicates a poor understanding of the strengths of focus groups. Groups are best for wide ranging discussions searching to discover what’s important in order to end up with impactful insights.

But aren’t Focus Groups an Artificial Environment?

This is a particularly pernicious argument. No research is conducted without interfering with the consumer’s life. So, we need to get past the false “common sense” of this criticism.

In any research of any type, the environment is a CHOICE to be made ahead of time and to be made with care. If we need to learn X, where can we best learn that? And, isn’t it the facilitator’s job to run a group in any environment? Be it a basement, a church, a home or an office, an expert moderator can make the participants comfortable in any setting.

There’s no inherent value for being in a person’s home. Sometimes intruding into their home can inhibit discussion. Other times it might help. Again: What is the purpose of the research and how will location affect THIS discussion?

A simple example. Suppose you are learning about kitchen habits and choose to focus on women who are the primary kitchen users in their home. Ethnographers will tell you to go into their homes. So, a recruiter arranges a date and time for an appointment to be in a “real consumer” kitchen for an interview. All sounds simple. 

Except… Most of these women will, no matter what the recruiter tells them, clean the entire house with heavy focus on the kitchen. They’ll also dress differently than they would and have different things on the counter than they would normally. They will also likely be nervous because the kitchen is often the center of a family’s activity so what happens there reflects on the entire family.

These women you want to hear from will likely be more comfortable in you invite them into a facility. If your goal is to investigate how they organize cupboards, then you might still go to their homes — but your research evaluation MUST include the degree to which you interrupted their lives and consider the error introduced by that interruption.

Environment is a tool to use in order to learn what we want. Sometimes a facility is a help, sometimes a hindrance, and most of the time a neutral factor.

The Unique Psychological Dynamics of Focus Groups

As research has become a very big industry, I have ended up being handed a lot of research which might be well executed (according to “best practices”) but is entirely uninspired — missing anything which really matters. This flavor of mediocrity is a huge danger for business. 

So, if you’ve seen some focus groups and thought “that was a bust” all I can offer is that I have seen those types of groups too (usually on tapes proudly sent to us by a client). 

Let’s assume you have an excellent team executing the work — like the work Carla and I have done so much of together. There are unique dynamics which focus groups can tap to find what you won’t find elsewhere.

But before we get there, remember that direct questions don’t get accurate answers. This is a fundamental research law companies forget — because execs “want direct answers.” It’s NOT the customer’s job to give those answers — it’s the job of research discover the truth instead of verbatim accepting “what people say.” And, if you demand direct answers, you’ll be entirely misled – you won’t find truth but only what customers think you want them to say.

Here are a few key values of focus groups:

People are often more comfortable talking about delicate subjects in focus groups

This may seem counter-intuitive but I don’t think it is. A one-on-one discussion with a stranger often causes us to close up — when talking about anything. Worse, when a stranger in an intimate situation (like our home or a one-on-one interview) asks us a direct and sensitive question, we tend to give them perfunctory answers — not the truth.

On the other hand, in group dynamics no one is querying the participant. Rather the group is discussing a topic and the participant will chime in with how they see it — revealing far more than they’d ever reveal in a one-on-one situation. 

This paper talks about the delicate subject advantage — after first confirming the “make a list” findings from the MIT study. As stated in the paper’s abstract:

…several types of sensitive and personal disclosures were more likely in a focus group setting

…some sensitive themes only occurred in the focus group context.

…No sensitive themes emerged exclusively, or more often in, an individual interview context.

Nothing lets us see inside the head of research participants like focus groups.

When I use focus groups, it’s because we need to understand how people think. While ethnography might show behavior, I haven’t seen it reveal psychology very well. One-on-one research is less likely to discover what matters than focus groups on a topic where this truth is important. Groups do this because a discussion among a group of people with a specific commonality on a shared topic reveals far more than any individual interview. 

Why worry about psychology — isn’t observing behavior enough? Not at all. As marketers we need to understand how to persuade, how to help lead customers to choose our products, how to evaluate and trade-off the product options we are considering, etc… Those questions are equally important with behavior.

Observing focus groups, we watch, listen, read body language and even add new information to the discussion in real time — so we’re able to thoroughly observe how different things impact people and their psychology.

To succeed at this, it is critical that we be disciplined listeners (that’s critical in any research). It’s our job to understand what’s said — not just stick with verbatim comments. 

Note that I do NOT allow verbatim transcriptions of groups. Verbatim transcriptions primarily allow people back at HQ with no understanding of the discussion choose to believe that the research says what THEY want it to say.

Focus groups allow participants to think

When we are digging, talk time isn’t the only thing — or most important thing. Using focus groups for the right topics we need participants to have time to think about hard questions. So, as one person talks, other participants think about what is being said or has been said and puts that up against their own sense. As a result, they articulate things we’d never hear in one-on-one’s or find in ethnographic work. Participants may, in fact, need an entire 2-hour group thinking about what’s being said before a particularly critical truth rises to the surface.

Companies can’t force people to think — it only happens when they are given time, stimulus and are part of a discussion. And, importantly, a group discussion prompts participants to recall things they might ordinarily forget if interviewed individually. One-on-one gives research subjects get very little think time and all their time in the interview is made more intense because they get 100% of the attention from an interviewer.  

Focus groups are brilliant when used with appropriate stimulus

As I noted above, the groups Carla Roberts and I do together rely heavily on stimulus to make discussion far more useful. Too many groups just ask people to “talk” and are wasted time.

Together, we spend days preparing the stimulus — but not word-smithing like copywriters. Instead, we develop concepts to present the important ideas. It’s my experience that customers think well in terms of ideas — and given ideas in stimulus we come to understand how the ideas impact people singly and together to ultimately arrive at a persuasive mechanism.

Using this technique, we also often come away from the research with discoveries we would never have arrived at on our own. With the ideas presented in ways that participants can work with, we come to understand how the elements fit together. It really is a phenomenal process.

Learn how customers perceive their behavior in focus groups

It’s quite trite (yet common) for people to say “consumers don’t do what they say”. Of course, not — they are human. This isn’t a great insight — although it’s an appropriate caution to corporate managers who would rather customers were entirely logical.

But it’s critically important to learn how customers PERCEIVE what they do. These perceptions give us the hooks to decide, for example, how product features create the product which sells best or which gives us opportunity for the most persuasive language in our marketing communication.

Focus groups are a learning medium

Consumer research should be focused on learning in order to (a) sell more profitable product or (b) avoid wasting money making or selling product that’s not profitable. Even engineering based customer research MUST have this ultimate goal.

Focus groups are brilliant in that they allow us to learn, adapt, and grow our discussions as we proceed through a sequence of sessions. When we use ideas as stimulus, we may learn some are unimportant and drop them

Ethnographic research can also offer this learning opportunity. Quant doesn’t and can’t. Dial groups (as just another example) are very weak for learning. They make the mistake of trusting behavior (number on a dial) and jumping to conclusions about what it means. I’ve written more about this here.

Weaknesses in the Ethnography Alternative

The primary research fad suggested as an alternative to focus groups seems to be ethnography. I don’t have any problem with ethnography — when used at the right time for the right issues. My experience with research started with ethnography and for that project it was of exceptional value. When conducted well and with awareness of the risks involved with invading a customer’s environment ethnography can be quite useful.

That said, I don’t see much honest discussion of Ethnographic research’s weaknesses — hence why I consider it to have “fad” status. Let me mention a couple weaknesses. 

The most common error in ethnography is an assumption that it finds unvarnished truth. Whether the result of your individual presence in a home or customer location or the result of installation of cameras or other recording devices, your act of observing will always change what the customer does. Always. This is similar to the quantum physics problem where the process of measurement changes what’s going on among the atomic elements.

The second error seems to be an assumption that “observed behavior is truer than hearing people talk.” In my experience, this is absolutely false. I’ve spent a career around direct marketing where practitioners say “what we can measure is all that matters.” Except, that’s not true. We quite often observe behavior we don’t like (“not enough people bought”). Given this behavior, we have to make changes to improve it. Guess what, behavior won’t tell us how to — since we have no idea WHY the behavior happened or what it means.

Observed behavior is one-dimensional. In an ethnographic environment you can see far more than a direct marketing campaign. But any time we focus ONLY on observation, we lose something. I’ve written more here. In an observation situation, we don’t know what we didn’t observe.

Embrace the Unique Powers of Focus Groups 

All companies, ad agencies, market researchers, and marketers will drive better customer sales and satisfaction by using the unique opportunity brought by focus groups and by ignoring the silliness of the anti-focus group fad we’re in right now.

In order to do that, it’s critical that those funding the research take tremendous care in hiring the right expert to execute the groups and keep expectations in line with the valid findings generated from the groups. 

Remember, also that this same care must go into any and all research — whether qualitative or quantitative, whether ethnography or focus groups. No research delivers the uniquely important insights which make a tremendous difference for a business without that care.

And, as a final note, please fight back with every research project against research mediocrity— those studies which carefully follow “best practices” without uncovering results that are truly important. The true value of research must be finding that which offers competitive advantage.

NOTE: Special thanks to Carla Roberts for her many reviews of this post throughout its writing.

©2019 Doug Garnett — All Rights Reserved

Through my company Protonik LLC based in Portland Oregon, I advise a select group of clients to drive success with better marketing of new and innovative products. I also work with clients attempting to bring new life to Shelf Potatoes or take their existing products to new markets. You can read more about these services and my unusual background (math, aerospace, supercomputers, consumer goods & national TV ads) at

Categories:   Business and Strategy, Consumer research, Marketing Research, Research & Attribution