Issue link: https://nebusinessmedia.uberflip.com/i/482857
Q U A L I TAT I V E R E S E A R C H C O N S U LTA N T S A S S O C I AT I O N 39 on key questions, such as how to mone- tize media in a digital environment with many competitors; how traditional quali- ty content companies stay healthy in the era of bloggers; user-generated content; ubiquitous digital photography; and low barriers to entry. Trying to anticipate the future keeps you on your toes, and in the current disrupted media world, the future is now. You might say that there has never been a greater need for research and insights. Right now we have more data than ever before, but we struggle to convert the data into meaning. Some of the problems are common to all types of research: how to mesh together online and offline, ran- dom and non-random, how to integrate disparate data sets to get to better infor- mation. In the media research context, this sometimes takes the form of the "attribution" problem. Who gets the credit for the sale? People are subjected to so many different commercial messages that everyone knows it is a mistake to credit only the last click. You would miss all the messages that prompted them to do a Google search in the first place. Years of prior brand impressions – some of which were measured in detail, although some were not – or something else could form an association between a brand and a particular set of attributes. The process of trying to figure out how to model and attribute the credit for the consumer action is a very interesting, dif- ficult, but important area of methodolog- ical innovation. It's one of many exam- ples of the gap between our growing abil- ity to gather data and our still limited ability to tame it. I go back and forth on the neurosci- ence methods. They frequently seem to be difficult to scale and to make sensible business decisions to use, versus simpler methods that can get you the same kinds of conclusions. It's like the old galvanic skin response or pulse monitors: you might be able to detect a change in behavior, but how do you actually know what that represents? Instead of sweaty palms, now it's the brain lighting up on some fMRI. What's the substantive mean- ing? We have better tools for seeing brain activity, but are still vexed by the problem of interpreting the results. We've done research for many years about how people do not give true accounts, forget certain things, or recon- struct their history without even knowing they're doing it. There are all kinds of ways in which our retrospective accounts have been shown to be unreliable, yet they're the only thing that we have had available. Our culture is biased in favor of passive measurement or direct market test, as opposed to the toolkit that we typically use in research. As in business in general, we in research are vulnerable to faddishness. We sometimes mistake tools for methods. We've got a new neuroscience tool and we think, "That's going to be the answer to all my problems." Instead of treating it as a tool, we treat it as THE guide to truth, and it's not to come up with a rea- sonable hypothesis about why these things are true. You still need that smart human being asking, "What does it mean and how can I evaluate that with other bits of evidence that may be aligned or may be contradictory?" Judy: How have your decisions on what qualitative methods to use changed? Where does qualitative fit in for you now? Scott: I usually recommend qualitative be in the mix, at least for editorial and consumer projects. The side of the busi- ness that involves making claims to advertisers really needs quantitative data. It's paradoxical; if you can't show it to them with numbers, they're usually not going to pay attention to the other stuff, but if you get a couple of consumers on video to illustrate what you're talking about in the numbers, then that's actually what they remember – because they're just human beings, too. It's like a mental shortcut, a heuristic. They can say, "Oh, I get it, she's articulating that," as long as they have some quantitative support. But if you make the claim by showing them just a handful of people, it's difficult to get past the skepticism, particularly in the really large agencies and consumer prod- ucts or financial services companies. If we're dealing with the luxury advertiser in Paris or Milan, like a fashion house, though, they tend to operate more intui- tively. They're with a company like Condé Nast because they can just sense that they belong in Vogue, that their product just looks good in there. Judy: Do you still recommend that quali- tative be part of the mix editorially? Scott: Yes, and we have better tools now. For a project coming along later this year, we will recruit a temporary online com- munity of women who are in this catego- ry, but aren't necessarily brand loyalists. The plan so far is to send them the maga- zine before and after the redesign and have them talk among themselves in the community as a way of monitoring it. That's a technique that we didn't have available a decade ago. To be honest, most of the online quali- tative techniques leave me feeling unsatis- fied. I get some information, but it isn't as rich as what I get when I actually drag myself out into the field and watch peo- ple. That's partly because I like to watch people. Sometimes they say one thing, but I'm watching their body language and "Particularly when I'm trying to get editors to think more concretely about who's in their market, you still really need to structure some direct engagement between them and their customers. For this, I prefer more old-fashioned high-touch approaches to qualitative."