The Warc Blog

The Warc Blog

Killing the conversation
Mythbuster, Les Binet and Sarah Carter, DDB

Les Binet and Sarah Carter of DDB get a little bit angry about some of the nonsense they hear around them… like the idea that focus groups have had their day.

Nearly 20 years ago, we were working on repositioning a global whisky brand. This required us to attend some qualitative research in New York and Los Angeles. Intensively trained in the art of conducting what we still quaintly called ‘group discussions’ at BMP, we were stunned at what confronted us.

In the UK, groups were convened in intimate surroundings in real people's houses, and were rarely attended by clients. A common-sense and trusting approach to recruitment and discussion guides prevailed, and, above all, time was protected to analyse and interpret fieldwork before the debrief.

In US ‘focus groups', things were very different. The research all took place in beige, faux boardroom settings – usually around a large table. Clients sat behind a one-way mirror, chatting loudly, ridiculing respondents’ ignorance of their brand and paying more attention to the takeaway menu than the proceedings. The moderator had little role, beyond putting a prepared list of questions in a set order to the bored respondents. After a quick chat with the moderator, the clients each went away with their own idea of the research findings.

Back then, this seemed a travesty – a pathetic distortion of the true principles of qualitative research. But 20 years on, this experience is increasingly becoming the norm.

Here are a few of the situations we've encountered recenty: crazy requirements to recruit respondents according to a complicated client segmentation involving lists of 30 statements and nine-point scales; weeks spent signing off a seven-page discussion guide, with numerous clients involved in changing odd words on it, so'everyone is aligned'; clients flying to the US for just two groups, arriving halfway through the first and chatting through the second; marketing teams spending weeks micro-managing every aspect of the research process before the fieldwork, but then instant debriefs routinely required after groups. That's before we even start on the issue of web streaming, where we have weary clients in their pyjamas logging in and out during research in the small hours on the other side of the world, busily adding comments to the text box and communicating with colleagues during the research.

It's almost de rigeur to bash focus groups these days: they're old-fashioned and you don't get anything new or different from them. The problem, though, is not with the focus groups, but with the way we use them.

When expertly practised, focus groups can provide a uniquely sensitive understanding of people's relationships with brands and communication. They are probably still the best means at our disposal of accessing what Nobel Prize winner Daniel Kahneman calls' System I’ thinking – the emotional, implicit, non-verbal thought processes, which neuroscience tells us drive most buying behaviour. The silences, the arm crossing, the raised eyebrow, the words not chosen, the laughter, the leaning forward or leaning back, the change in group energy level – all these can be far more potent indications of what is really going on in people's minds than the words they use.

The trouble is that focus groups seem to be increasingly interpreted now by clients at Kahneman's ‘System 2’ level – the realm of the verbal, logical and post-rationalised. One step removed from the respondents on the other side of the glass, clients tend to concentrate solely on the verbal content of the discussion. Just ask yourself this question. If you were being tried by a jury – would you rather they were in the room with you, or in their pyjamas at 2am watching a small screen on the other side of the world?

Through our industry's well-intentioned effort to make focus groups more ‘rigorous’ and 'scientific', we have neutered and distorted them so they no longer deliver the thing they were designed for- sensitive and empathic understanding of other people's perspectives. As a result, they are at best much less useful, and at worst downright misleading.

So if your focus groups no longer work for you, ask yourself this: is it the technique, or is it the way you are using them?

Subjects: Consumers

01 February 2011 12:01

There are 8 comments on this blog

all so true. I miss the old days
richard m. 04 February 2011 at 5:41pm
Well said and about time. Wonderful tool in the right hands. Routinely abused by clients (and even misinformed agency folk) who treat it as some kind of semi-quant exercise, and fail to realise that their endless interventions ruin what you're trying to create. No coincidence that same clients will have a mechanical and over-rational approach to comms.
Adrian L. 06 February 2011 at 9:51am
At Delta, we have a service called "Turbo Focus Groups" that we feel provides a more quantitive and actionable approach to analyzing results. Starting with an video capture of the group, we do about a 10 to 15 minute "hilights" block covering the key points where the client wishes to know more about how prevelant are the issues discussed, or how relevant are the solutions proposed.  This is shown as a streaming video to a sizable (say 800) person internet panel, and their reactions (like/dislike, interest/disinterest)  captured on a 1-10 scale controlled by mouse pointer.  This gives us a basis for quantifying the ranges of response, gaging the degree of difference among key segments, and in general allows  focus group results  to be analyzed in a quantative as well as qualitive manner.

We also use a similar method to analyze differential response to tv commercial elements by segments as early as in storyboard form.
John T. 24 February 2011 at 3:48pm
Excellent, thanks John. And I'm sure they're very reasonably priced too. Have you actually read what we wrote?
James H. 24 February 2011 at 4:04pm

About time.

It saddens me the way that people are so quick to slag off the group discussion (sorry, I refuse to call it a focus group!). 

Qualitative research is about talking to, observing and listening to our consumers.  There are many ways to do this, and the group discussion is but one way of doing it.

Used in the right way, there's nothing to beat a group.  All too often, though, it's done badly and used badly.  How many agencies/researchers actually analyse the output of their groups, as opposed to a quick bit of team brain storming?

How many clients go along to one or two groups, talk/eat/hold meetings during them and then think they have the answers that they need?  I have worked with one client (who shall remain nameless) who told me they didn't need any analysis/debrief - they had what they needed after watching two groups.

But who is to blame for this?

Alison S. 09 March 2011 at 12:40pm
I have experienced that " we dont need any analysis  or debrief " comment too. You are right to raise the issue of who is to blame. I guess we all have a collective responsibility to educate re what is good and bad qual. i think many clients dont really know and have no criteria by which to judge.
sarah c. 09 March 2011 at 3:29pm
Great comments and article. Now that we have decided to really respond to the criticism, I think we are hearing some great discussions. Unfortunately, a lot of us are preaching to the choir.

At times it seems the clients who demand and value high quality research are a rare treasure -- perhaps more reflective of the forces at work in the corporation today than the dynamics of research alone.

Chris Kann and I wrote another take on this topic recently, published here:
Susan A. 18 March 2011 at 1:37pm
@Susan A.: Amen, sister! There is a lot of preaching to the choir going on. (Your article was a fine one, by the way).

What I'd like to see us do is spend half the energy we seem to spend talking to each other about this, educating clients. So let's get out there, people, and demonstrate the value of listening for meaning, instead of content!
Megann W. 18 March 2011 at 1:48pm
Comments IconAdd your comment here:

Blog Search


  • 2016
    • October (11)
    • September (11)
    • August (18)
    • July (11)
    • June (16)
    • May (10)
    • April (17)
    • March (16)
    • February (13)
    • January (12)
  • 2015
  • 2014
  • 2013
  • 2012
  • 2011
  • 2010