The success or failure of the insights industry is entirely dependent on the quality of our relationships with the people who complete our surveys. At Maru/Matchbox, we are committed to treating our survey respondents like people—people willing to share their time, opinions and experiences with us.
We recently conducted video interviews with members of our Maru Voice UK community, and asked them what they like and dislike about doing surveys. We also solicited suggestions for improving the survey experience. The results reveal a need to radically change how the industry conducts research projects and how it treats respondents.
Curiosity and Contributing
Many people complete surveys because they want to make their voice heard. “I like sharing my opinions” one woman told us. They feel that by sharing their perspective they can make a difference. Asked why he did surveys. one gentleman said, “most of it is because I think I’m making a difference in what we are doing….” Another woman said her reasons for doing surveys were twofold: “One, it’s because if companies, manufacturers and stuff don’t know what the customer or the people know or want or think about their products, how are they going to improve them, or change them, or make them better, or if they need to change them at all? Second reason I do it is because 9 times out of 10 I just find it quite interesting….”
Receiving honorariums is also a reason to do surveys, but as we have seen from other research on survey incentives it is a relatively weak motivator that is more of a symbolic gesture of goodwill and respect than a payment for time spent. These UK findings are consistent with what we have seen in the U.S. and Canada.
One theme that popped up again and again is the dislike of doing long screening surveys only to be told you do not qualify. We know that, in a river sample environment, screening questions are typically repeated over and over before a person qualifies for a survey—if they ever do. It’s an excruciating survey design experience that I lived to write about in Bad Respondent Experiences are Killing the Insights Industry.
At Maru/Matchbox, our survey participants are spared this frustrating and exploitative experience. Our survey community members are richly-profiled, and we use that knowledge to target the right people with the right survey. We don’t need to ask things like demographics because we already have that information and can simply import it—which we have proven to provide accurate information from known respondents.
Long Surveys and Frustrating Questions
Other frustrations people expressed included being asked questions that did not apply, and long, overly detailed surveys. As one woman explained “They ask me, so what kind of products do you use? What shops have you used before? And it goes on from like Tesco, Lidl, Amazon, H&M, Argos, whatever. It’s like endless lists that I’ve been to. I’ve shopped everywhere. Like, who hasn’t? Who hasn’t bought one thing in Argos, and then hasn’t been back for 2 years?” Here the question is both too long and not applicable. How accurate can information about a single store visit from 2 years ago be?
The frustration with disqualifications, the river sample experience, and long, repetitive and overly detailed surveys that we see in the UK, have been echoed across the pond in the U.S. and Canada.
Annie Pettit has written a valuable and very readable book about writing surveys. The title of the book People Aren’t Robots: A practical guide to the psychology and technique of questionnaire design, speaks volumes. In an email exchange for my book The Insights Revolution: Questioning Everything, I asked her thoughts on the way surveys are written today. She marveled at how we can expect things of respondents that we can’t do ourselves.
“We forget what we’ve bought, we misremember where we’ve shopped, and we rationalize that we buy things because we need them, not because we want them. But when it comes time to communicate with people participating in our research, we instantly block out our own illogical behaviors and expect them to remember precise SKUs, stores, dates and times, and so many other meticulous details about every purchase decision they’ve ever made. That is the definition of illogical.”
“We expect people to not get tired, bored, forgetful, or distracted when participating in research about topics that are only interesting to the brand manager,” Pettit continued. “We expect people answering questionnaires to choose valid answers when none of the answer options apply to them and provide diverse responses to grid questions that demand straight-lining. And rather than getting annoyed at our own failure to plan for and design research that accounts for the illogical creature that is the human being, we instead call people liars, cheaters, and fraudsters. Market researchers are supposed to understand and be empathetic towards the illogical, forgetful, perfectly imperfect human being.”
I think the people we interviewed would say “Amen” to that.
Reduce Repetition in The Survey Research Experience
We asked people for suggestions for ways to improve survey user experience to keep respondents engaged. There was a consensus that people want surveys to be more interesting, and less boring and repetitive. And they have a point. Too often researchers ask the same question from multiple angles, fearful that if they might miss something. This is a terrible experience for survey takers, who respond to the repetition by becoming less engaged, decreasing the survey response rate and data quality.
What’s more, this repetition is unhelpful in many ways when it comes to analysis. First, it gives us no new information. We get basically the same answer to basically the same question, over and over again—which helps give research the reputation of being boring. It also focuses researchers on the task of reporting, at the expense of time spent thinking about what the research means.
Let’s consider this example, from a test of a potential television show. A targeted audience saw the show and then were asked a series of questions, including these:
- Do you think this show is excellent, very good, good, fair, poor?
- How much do you want to watch more episodes of this show?
- Would you recommend this show to a friend?
- How entertaining do you think this show is?
- Would you want to meet the characters on this show?
- Do the stories in this show keep you interested in seeing what happens next?
- Do you like what this show is about?
- If you were the boss of a TV channel would you put this show on your channel?
The answer to all these questions is essentially the same because they are measuring one underlying thing: whether people like the show or not. A factor analysis confirmed what a sensible reading of the questions suggests. You don’t need to ask the same question over and over again, in slightly different ways. The only question that is needed is the first one. The rest are just noise that alienates the people who answer our surveys. All the other questions are what Annie might call “only interesting to the brand manager.” But the brand manager will be disappointed anyways because all the answers don’t reveal anything more than people like or dislike the show.
Let’s stop playing overly safe by asking the same question over and over. Let’s craft a few good high-quality questions and then trust the answer. Make this your mantra: when in doubt, cut it out.
We need to heed the voice of the people who do our surveys. We need to avoid using river sampling and use well-profiled community members instead. We need to tell survey respondents how much their opinion is valued and how it will help us make decisions that will serve them better. And we need to avoid repetitive questions that alienate those respondents while giving us no additional information.
We must heed the voice of the people because without them the insights industry is over.
We’re in the midst of an insights revolution. Vive la révolution!