Tracking Success in Online Ads’ Brave New World

Monetizing social media

SANTA MONICA, Calif. — Few things thrill online advertisers more than metrics — those magic statistics that, at least in theory, enable them to know exactly who is viewing their ads and whether those viewers successfully made a purchase, clicked on the ad or signed up for a future mailing.

But what worked during the heyday of static banner ads hasn’t necessarily translated to today’s online advertising. Instead, the industry has struggled to come up with metrics for tracking newer forms of marketing that use video, social media and user-generated content platforms. And with no way to measure the success of their campaigns, businesses have no clear guidance on whether an ad was a success.

Figuring out the answer now may be more critical than ever. Advertising sans metrics makes brand owners very nervous, so measurable metrics “couldn’t be more important than in this current economic environment,” according to Mark Ghuneim, CEO of digital agency Wiredset, who moderated a panel discussion here on the topic at the Digital Hollywood conference in Santa Monica, Calif.

Panelists agreed the industry needs to better understand in area in particular: “engagement,” which they described as when the consumer enters into a relationship with the products and brands they use. It’s not enough to simply buy stuff anymore; ideally, consumers buy into the whole brand experience. In some successful cases, like Apple or Harley Davidson, consumers see a brand as an extension of their lifestyle.

One problem is that most consumers have little interest in engaging deeply on a profound level with every product that they use. Worse, panelists seemed at a loss in terms of how to better grasp or measure engagement, which is a notoriously elusive concept, said Charley Shoemaker, director of video measurement products at Nielsen Online.

Lessons from the past

This craving for measurable statistics led to an overreliance on “click through” back in the formative years of the Internet — after all, what could be a better predictor of the success of an ad campaign than counting the people who saw an ad and responded to the call to action by clicking on it?

Well, as it turned out, simply clicking or not clicking on an ad is not a strong indication of the success of a marketing effort. People may not need the product being advertised at that moment, but may remember the ad — and the product — when they do need it, later. Klutzes may click accidentally. People may click through out of boredom but never buy the product.

These days, advertisers know they need to look beyond the click. But that might be easier said than done. For one reason, panelists said culling together actionable metrics from today’s fragmented media market proved difficult for many advertisers.

The general rule of thumb is that statistics culled from a large group will provide more accurate information than those culled from a smaller number of people. But if you have thousands of Web sites and channels and media outlets, each appealing to ever smaller groups of people, you have to work with much smaller sample sets, which may or may not skew the across-the-board accuracy of the statistics,” said Ken Papagan, president and chief strategy officer at Rentrak, an audience measurement firm.

Others thought they had worked out a solution to getting granular data.

“People naturally segment themselves. With regard to video, we correlate video consumption within each segment, and when we start to see that certain types of segments favor certain video, we know what to expect,” said Ryan Burke, managing director of sales and business development at Compete, an online marketing research firm.

“It’s sort of like a collaborative filter,” Burke said. “We don’t ignore metadata and transcription data. We layer this on when coming up with predictive advice — If they like ‘X’ video, they’ll like this one. We’re doing this for Fox TV Web sites now.”

Marketers are also trying to track the “velocity,” as Wiredset’s Ghuneim described it, of what people are saying to each other on Twitter, blogs and forums, to see how fast information moves across social media and what impact fast communications have on advertising.

Do ads have more of an effect on mobile devices? Does Twittering have a measurable benefit for a company? Are blogs devoted to a single topic working? No one really seems to know yet.

Other media create other hurdles to accurately quantifying brand engagement. For example, interest in measuring advertising success on peer-to-peer (P2P) file-sharing networks is fading, due to lack of cooperation from the owners of the sites, according to Shoemaker.

But marketers continue to be enthralled by the idea of measuring consumers’ conversations online to find out how people really feel about their brand.

“I think there’s tremendous value in monitoring all of these forms of discourse, whether it’s blogs or other social media,” Rentrak’s Papagan said. “What is going on in the cosmic consciousness of America is what we’ve been trying to understand for hundreds of years. We need to understand what people are thinking so that we can bring them the products that suit their needs.”

If marketers can’t measure what consumers are thinking, what will happen? According to Nielsen Online’s Shoemaker, if we don’t have stats, then we don’t have advertisers.

And if we don’t have advertisers, the profitability of the Internet will come tumbling down, Shoemaker said — just “like a house of cards, as it did 10 years ago.”

News Around the Web