BATAVIA, Ohio (AdAge.com) -- Replacing "asking" with "listening" has been a hot topic at market-research conferences for the past couple of years. But now some researchers are finally doing more than talking -- they're taking steps toward replacing surveys with web tracking.
Already online-buzz tracking has gone from exotic specialty to growth industry, with entrants popping up like mushrooms after a heavy rain on the promise that marketers can monitor brand health by observing and analyzing people's spontaneous comments in social media and other web venues. At forums such as the recent Advertising Research Foundation Re:Think conference in New York, market researchers talk increasingly favorably about such "listening" approaches.
de No marketer yet appears to have taken the plunge to online listening only, and even some longtime advocates of web tracking, such as Nielsen Online Exec VP Pete Blackshaw, are reluctant to suggest marketers ditch any of their tracking surveys just yet.
But some are suggesting they can be a dominant form of research.
"The way a lot of people are using social media right now is to help explain survey results," said TNS Chief Research Officer Larry Friedman. "What we're arguing is that you need to think about it the other way around" by using surveys instead as the supplement to expand on trends identified in various forms of web tracking. He said it's far too early to throw out all tracking surveys, but that they can be reduced in frequency and length by substituting web tracking.
Watching how you browse
It's not just about observing conversations in social media, Mr. Friedman said, but tracking people's online behavior in other ways, such as visits to websites and search queries. To that end, he put together a study comprising web traffic, Google-search volume, sentiment of online buzz and tracking surveys from TNS and other sources to show how Tiger Woods' scandal affected his popularity and the brands he endorsed.
One advantage of relying more on web tracking and less on surveys, he said, is to counter a growing trend toward longer, more-involved surveys time-pressed, survey-weary consumers often hate to complete.
Of course, TNS owns Cymfony, an online-buzz-tracking service, and tracks web traffic via its Compete service, so it has skin in the online-tracking game. But Mr. Friedman also employed data from other providers, including syndicated ad tester Ace Metrix and Google Trends data.
A better indication that the industry is ready to embrace web tracking for broader use comes from Kim Dedeker, former market-research chief at Procter & Gamble Co. and now chair of WPP's Kantar Americas, of which TNS is part, but of which Millward Brown, one of the leading purveyors of survey-based tracking research, is also part.
Five years ago, while at P&G, Ms. Dedeker expressed skepticism about blogs as research tools, noting that bloggers weren't really representative of typical P&G consumers. But asked last week whether it might be possible for marketers to replace tracking research with online sentiment monitoring, she said, "It's very possible. We need to find better ways to enable it."
Tracking surveys remain popular with marketers because they offer the comfort of decades of validation, she said. "But we can't keep doing things the same way forever."
The fundamental problem with tracking surveys, she said, is relying on responses from people who often don't care about the questions beyond the compensation they're getting, she said.
The passion, or even lower-level caring, that survey research tends to miss is one thing online buzz may be able to capture, in some ways better even than ordinary word-of-mouth.
In an ARF presentation, Jonah Berger, a Wharton marketing professor, noted the reason 90% of word-of-mouth about brands comes offline rather than online is because offline conversation tends to be more casual, while people usually have a higher threshold of being interested and emotionally involved in a subject to make a comment online. "What we see with online content is more emotion," Mr. Berger said. "Content that makes us feel emotions makes us want to go out and talk to other people."
Similar observations have led David Wiesenfeld, a longtime researcher who for years relied on tracking surveys both on the client and vendor side, to believe tracking surveys miss too much of consumers' passion and complexity. Mr. Wiesenfeld is VP-marketing solutions for Nielsen Online now, so he has a vested stake in marketers moving to web tracking. But he's also a longtime researcher with roots in survey research with Ocean Spray and Nielsen's Bases concept-testing service.
"If anybody ever listened to a telephone interview or watched a consumer do an online survey, I think they'd feel a lot better about [web tracking]," he said. What also gave him doubts was seeing survey-tracking results that didn't correspond to market movement during his time at Ocean Spray, which he left for Nielsen in 2005.
"Things that are more emotional, things that are more complex ideas than it tastes good or doesn't, surveys don't easily express that," he said. "If your advertising isn't working enough that people are spontaneously talking about it, is it really working? Probably not."
Yes, but is it valid?
But while web tracking is cheaper, faster and, at least according to Mr. Wiesenfeld, better than survey tracking, he said researchers still haven't convinced themselves or their bosses that it's better.
What's missing is statistical validation of web tracking. That validation may be coming, if ever so slowly.
Zach Vetter, VP-research operations at Toronto-based OnResearch, said online-buzz monitoring's position today is "akin to where online research was a decade ago. People were interested in it but saw it as complementary. I think we're seeing the same sort of phenomenon with social media. Marketers are interested but want to know if it's valid."
To that end, OnResearch last month commissioned a parallel study of 75 brands using a survey of 524 consumers side by side with tracking 200,000 blog comments about the same brands. What it found, Mr. Vetter said, was that the "net promoter score" -- essentially the number of survey respondents with positive outlooks on a brand minus those with negative outlooks -- correlated closely with the "net blog score," OnResearch created, which subtracts the number of negative blog comments from the number of positive ones.
Still, there's considerable need for analysis, and plenty of evidence that online comments, particularly the occasional Twit-storms of outrage over a marketing campaign or mishap, don't always reflect sentiments of the broader population.
Rob Schwartz, chief creative officer of TBWA/Chiat Day, Playa Del Rey, Calif., is among those already listening closely. He has Tweetdeck on his Mac loaded with standard searches on some of his agency's big campaigns, such as the recent Winter Olympics effort for Visa.
Such monitoring is "ever present," he said in an interview earlier this month, "at least at this agency. ...It certainly gives you the feedback you're looking for -- and not looking for -- faster than ever. ... If you see a lot of negative stuff, you say, well, what are we going to do, because they're not liking this idea. "
Sometimes it seems most people on Twitter are marketers, but Mr. Schwartz said it's really more that average people have gotten used to talking about ads. "The social web has opened up people to realize, 'Hey, if I complain enough, they'll take a spot off the air. Or maybe I can give them a free idea,'" Mr. Schwartz said.