Groups can work best when they don’t focus! The name ‘focus groups’ can be horribly misleading. Focus is fundamentally the wrong approach towards any discussion groups. Indeed, the methodology should be the loosest and least structured of any research you undertake. The value lies in not leading, but listening, following and shaping. And the further you get from the product at the start, the better. Successful magazines understand that their product may be a long way down the list of their readers’ priorities, so time invested in understanding behaviour, interests and motivations of consumers is never wasted. To help this, there are two important things to remember when briefing a researcher or considering a discussion guide for approval:
1. Fewer objectives reaps rewards.
2. The shorter the discussion guide, the better. A long discussion guide defeats any serious attempt to provide depth and context to your understanding of the consumer and their relationship with your magazine. A good guide will allow time for unforeseen discussions that can unlock hitherto unknown aspects of your brand. Allowing the moderator latitude in the way the guide is used on the day is also equally important.
Recruitment – the Cinderella of research
One of the most vital parts of the research is the recruitment – both how the sample is constructed and the screening process itself. Yet, few ever get to meet those unsung heroes of the research world who put their effort into getting people to give up their time and attend. Beware too, the ‘professional respondent’ who has attended numerous groups – always on hand with a quotable quote. Check that the recruitment guide screens out anyone who has attended a group in the recent past. (I specify a maximum of two groups ever attended and none in the past 12 months). This can be difficult in over-researched and small markets like IT directors of large companies, but it’s worth spending some time considering the recruitment script for people like this and providing convincing reasons as to why they should attend. A financial incentive is not the be-all and end-all. Networking with peers or spelling out some of the issues that will be discussed can also be important motivators.
Much can be gained from the construction of a careful matrix of respondent attributes (frequency of reading, age, organisation / sector type etc). But, remember that the more attributes there are, the more groups will be needed. It is never a good idea to do just one group for an important attribute, because there’ll be no benchmark to judge the findings by. Each group has its own dynamic, and it’s crucial that the findings based on these attributes are robust.
It’s equally important to take into account the softer issues in recruitment. How often have you walked away from a group feeling that they just hadn’t opened up? It’s a frustrating experience and one that your research agency will work hard on minimising. It starts with the screener. There should always be one or two questions in this recruitment guide that are open-ended. Their aim is solely to assess the respondent’s aptitude for group work. Are they articulate? Chatty? Will they dominate a group? What level of understanding do they display?
How groups work
This is a complex area and there’s little I can do in one article to do this justice. George Silverman, a US researcher of many years’ experience, has written extensively on this subject. Essentially, group discussions work at five levels.
| The 5 levels | |
| 1 | Superficial |
| 2 | Games |
| 3 | Protective |
| 4 | Authentic |
| 5 | Core |
| (With acknowledgement to G Silverman: Getting to the right psychological level in your focus groups.) | |
* The superficial level
This is the basic lubrication of all conversations – socialising before a group starts or before the group introductions. It would be wrong to see this superficiality as unimportant though. Early arrivals for groups will strike up conversations with each other; many of the questions searching for things in common; which organisation or sector they work in; people they know in similar places. It’s important because the friendlier the group, the more likely it is to work well.
* The games level
This is where the camouflage starts, and where literal interpretation becomes so dangerous. It’s the territory of one-upmanship; of grandstanding. Everyone does it. And much of it can be subconscious. How many times have you seen these characters in your groups? The Knowing Cynic. The Lad. The Expert. The Dominator. These are all games people play and you can never screen them out, but the group moderator needs to bear two things in mind. Move them up from this level and resist the easy option of taking what they say at face value.
* The protective level
In any group, there will come a time when the discussion moves away from the rational and logical towards the emotional or attitudinal. People get defensive – either because they worry about what other people might think or because they aren’t sure about what they really think themselves. Often, this defensiveness will be cloaked by a seemingly rational response. It’s all too easy to be fooled or to accept the comment and move on. And this is indeed just what happens for a large number of groups. Why? Often, it’s simply because the moderator is only half way through the discussion guide and has to cover too much ground. Time is the researcher’s number one enemy and allowances have to be built in. When it is, the moderator can probe further and get the group more comfortable in elaborating their thoughts on some of these issues. If not, these smokescreens can resurface un-interpreted as verbatim comments in the debrief – a sure sign that your research has failed to dig underneath for the underlying motive.
* The authentic level
This is when you know a moderator is delivering the goods. It’s when a group becomes relaxed enough to get beyond the grandstanding and starts to be open. Disarming humour is often an excellent tool in the researcher’s armoury to get the group to this point. You’ll start recognising candid comments – quite often in situations where respondents try to articulate something that they may never have thought much about before that moment.
* The core level
The core level is reached when talking about the most basic and powerful of motivations involving social convention, ethical tenets and the inner self. People are often reluctant, or feel unable to participate at this level, and getting to a conversation this deep or of this profundity is unlikely – particularly with a long discussion guide. Yet, this is what Procter & Gamble, and others, spend millions of pounds finding out about: not just the values associated with brands, but what lies behind the basic need to be clean.
Analysing the research
The report needs to compartmentalise the factual record of results from the analysis and interpretation. Analysis is not about faithfully recording and regurgitating the research transcripts. Indeed, the best analysis can often start whilst the groups are still in progress. This analysis is all about the recognition of the value of a particular remark. It is the point at which a good moderator will abandon the discussion guide and follow the new direction. It might help if I gave an example.
I was running some groups amongst dairy farmers on behalf of a consultancy involved in cattle breeding. I was getting good signals from the groups in terms of product need. There was a clear argument being developed as to why farmers saw this area as being fundamentally important – all good rational stuff, suggesting that the product was in the right market to be a goer. The mood changed, though, when we started broaching the subject of who would use it. Fingers were pointed in all directions, but everyone felt that it was better for someone else, not them. It was a simple projection technique which provided the breakthrough. Central to this was the sentiment that a farmer knows his herd better than anyone else. It would be a mark of failure if advice was sought – the analogy of a garage mechanic who asked someone else to service their car, was used.
That was the point when remarks made at the original briefing session fell into place. Despite very strong renewal rates amongst a small core of users, new subscriptions were proving hard to come by. This fundamental barrier was simply not being addressed in either the advertising or the sales pitch. With this new learning, a completely new approach was devised based on testimonials using well-known managers of dairy herds. At the core of this was the peer message that successful farmers seek advice in this crucial area.
