Content Writing

10 Keys to More Effective Qualitative Research

Writer / Author Christopher Vickers, Atlanta-based Brand, Advertising + Content Marketing & Communications Strategist & Creator

Qualitative research (focus groups, triads, 1-on-1s etc.) used to gather feedback and gauge the effectiveness of communications (e.g., new brand identity systems, packaging, advertising campaigns etc.) can be a bit of a minefield. Client, agency and even moderators can have preconceived notions about the concepts and favorites, usually not the same. And even the smartest and most self-aware consumers have difficulty articulating a useful response, and can be influenced by group-think or individuals who are more talkative and persuasive.

So it is with over over 20 years of experience developing, observing, interpreting and helping summarize qualitative research for scores of companies these 10 tips - some fundamental, some nuanced - emerge, which corporate and agency folk alike, across departments, find insightful and practical.

• Use an expert - Ask around, understand the researcher’s expertise. If, for example, you want some creative development research then go to a company which specializes in communications research - probably not the same people that ran the segmentation study. It’s experience with the type of research not the industry or vertical that really matters.

• Understand the idea - As well as the basics like agreeing clear objectives it can be useful to get everyone to write down their definition of the creative idea being researched beforehand. Usually an interesting exercise in itself. For example, ‘Is the dancing robot in the X5 ad a way of dramatizing the fun and lively personality of the car? Or a demonstration of its versatility as a hatchback? Is it a means of reaching out to 20-year olds who remember Transformers or is it just a funky track and an amusing piece of film to catch the eye and stick in the mind?

• Don’t forget the ‘off the record’ briefing - Your researcher is potentially a great ally and a second ‘political’ briefing is a great way of getting them on board. Give them the low down on the different members of the team - the covert agendas, the relevant brand language, the important bits on that mystical brand onion chart. The more they know about the issues, the less likely they are to put their foot in something smelly (“Don’t mention the awful packaging”). The whole process will run more smoothly and you'll be able to relax a little.

• Don’t ask their ‘opinion’ on the work - You can ask them about their opinion of the work but don’t expect an answer. Expect to be misguided with a fuzzy, positive but non-committal answer. Otherwise when they come back and say the idea ‘underperformed’ you’ll suspect an ingoing subjective bias against it. Besides if they could accurately predict which path will research best it would take all the surprise out of the job and make life boring.

• Qualitative samples are different - When it comes to sample design remember it’s not quant. Qual. samples aren’t about replicating the entire market. If you’re thinking more than 8 groups, take a step back and check whether you’re over-complicating the sample criteria. An overly convoluted sample risks weighing down the whole project with negative energy. Do you really need to recruit 33 37-year old women, regular users of brand X but semi-lapsed users of brand Y in Des Moines who have DirecTV and agree that “a real man can down several pints at one sitting”?

• Likeability is (probably) not the issue - If your researcher kills the ad idea at the debrief by saying that respondents didn’t like it, then promise yourself that you will never use them again. Who cares if respondents didn’t like it? Why didn’t they like it, and which aspects - fundamental and strategic or fixable and executional? And does likeability even matter in this vertical? Actually, you can probably salvage something at this stage by really grilling the researcher on their results. Even reportage researchers can, with a bit of help, admit the potential of an idea.

• Mind the stimulus - Spend twice as much time thinking about stimulus as you do about the sample or methodology. A good researcher can work with bad stimulus but it’s an uphill struggle. Give your ideas the best chance by making the stimulus development a priority. Concentrate on expressing the core idea and remember there’s no need for it to be consistent across routes. The looser the storyboard drawings the better – it stops respondents getting hung up on execution and gives them a chance to use their imaginations. Remember the illustrators will typically depict women as fantasy babes housewives will resent. Fudge the timing. Say the deadline is 24 hours earlier than it really is. Planners often see the storyboards too late and hand them over, sheepishly apologizing for the misleading aspects of the drawings. Get in there earlier and have a say.

• Beware oversell of the stimulus - Creatives’ understandable tendency will be to sell the idea rather than describe it. Beware narratives which spell out the strategy “we can clearly see she is thinking how effective and reliable the brand is”. No we can’t. And avoid purple prose. “We open on the most amazing mountain road you have ever seen. A stunningly beautiful car glides effortlessly…”. Not terribly useful if you’re trying to research what the ad can truly communicate. It can either seduce people so they say, “Yeah it sounds really good” and just play back the descriptors from the narrative, or it can wind them up. To get more out of the research keep it simpler and they can use their imaginations, develop more engagement, ownership and buy in. (The same way the best ads work.)

• Go to the groups - Do you really want to travel all the way out to the groups in a remote location on a wet Wednesday night? Yes you do. It may be inconvenient in the short-term but it will make you the best informed person on the team for months to come. Afterwards you’ll drop anecdotes about how people described the concepts that will make everyone feel you’ve got the target figured out.

• Try to manage the inevitable selective listening - Inviting the entire team to come along to view groups is a great way of keeping everyone on board and happy - the vast majority won’t turn up anyway. But don’t forget to brief everyone who you invite, beforehand. Remind them of those basics - the objectives of the research, the fact that respondents don’t often say what they mean or mean what they say. Clients can be great at selective listening – try and nip it in the bud when you’re sitting with them behind the mirror. If they walk out of the viewing room with a fixed idea, it can be hard to change. It is tense and claustrophobic back there and you need some release but keep the noise down. They can hear you laughing. Oh and if you’re sending several notes to the moderator under the door . . . stop.