Clients Try To Manipulate 'Unpredictable' Viral Buzz

March 19, 2007
by Brian Morrissey

When "Dove Evolution" was uploaded to YouTube last October by Ogilvy & Mather in Toronto, it took off like wildfire, drawing over 2.5 million views for Dove and several imitations. It was an instant viral hit that mixed creative firepower and Internet serendipity.

Then there's the seven-minute film by Leaving Las Vegas director Mike Figgis of Kate Moss in her underwear for Agent Provocateur, a lingerie maker that had what would appear to be the recipe for a viral sensation. But it was viewed fewer than 75,000 times in the three months after it was uploaded last September.

The fates of the two efforts show the predicament facing marketers who vie for consumer attention with campaigns they hope will go viral. Despite the populist nature of viral campaigns, companies are increasingly seeking assurances that their efforts will receive a certain level of exposure and ultimately converge with overall campaign objectives. The result is that some clients and agencies are seeding links on video-sharing sites and blogs, even trying to manipulate YouTube rankings to gain visibility-all in an effort to give creative a fighting chance to rise above the clutter.

Take Unilever, which in January enlisted the help of London-based "viral seeding" agency The Seventh Chamber to turn its Vaseline "Keep Skin Amazing" TV spot into a viral hit. By buying links on video-posting sites, leaving comments on forums and blogs, and other measures, The Seventh Chamber delivered 2.5 million views of the video in two months, more than double the number it guaranteed the client. All told, Unilever earmarked $100,000 for the effort, according to Richard Spalding, Seventh Chamber CEO.

"You can't expect something to randomly go viral anymore, even if it's good," he said.

Other marketers are turning to blogs to spread the word. For the launch of its all-in-one printer with cheap replacement ink cartridges, Eastman Kodak is running a 10-month viral campaign around the notion that printer ink is ridiculously expensive. To promote the campaign's YouTube video episodes of a mock ink-focused talk show hosted by ink-obsessed rubes Nathan and Max, Animax Entertainment and The Concept Studio devised a seeding plan that e-mailed links to small-business blogs and even links left in comments on the site. It has also identified less obvious groups that might like the offbeat comedy of Nathan and Max, such as sites by "scrappers," scrapbook fans.

"It's not something we can just create, put out there and hope it will work," said Ted Mandelkorn, evp of The Concept Studio. "That won't fly."

The move to bring a measure of predictability to the still-unpredictable world of viral marketing is being driven by clients trying to balance the risks inherent in a new marketing medium with the need to prove return on investment, said agency executives.

There is still debate over what constitutes a true viral campaign, as some feel it's one that happens organically, not planned out like a regular ad push. But clients continue to ask their agencies for efforts that tap the social media zeitgeist to build buzz. And they are spending more money on such "pull" marketing: the typical budget for a viral video has gone from $50,000 or less six months ago to sometimes topping $200,000 today, according to sources.

"There's no substitute to listening to consumers and doing that through analytics and analysis," said Jim Calhoun, CEO of Popular Media, a viral marketing firm that has clients like Red Envelope and eToys. "Otherwise, you're just guessing."

Denmark-based GoViral offers a guaranteed number of views or visitors, typically about 300,000 for a U.S.-focused campaign. The company uses a network of niche video sites and other destinations to get clients' content in front of influentials likely to spread it, using both paid and unpaid methods. For Nissan, it helped its agencies TBWA and Duke Interactive seed viral films ahead of this spring's Qashqai launch on sites popular with the car, sports and skater communities Videos are embedded with software to make it easily tracked, with the Nissan virals attracting 12.7 million views, said Jimmy Maymann, GoViral's chairman.

"We're trying to make it as scientific as possible," he said. "We want to make sure the clients get some bang for the bucks.'

Tactics vary when it comes to delivering on the promised views. At least one seeding agency tells clients it can land their videos on the list of YouTube's "most viewed" clips of the day, which is sure to drive interest. A popular approach is buying links on video-posting and youth-focused sites like MetaCafe and The advantage is the ad videos are mixed in with regular content.

Kontraband, a video-posting site in the United Kingdom, used its popularity with advertisers looking for a viral boost to spin off The Seventh Chamber. Thanks to Kontraband's large audience, The Seventh Chamber network purports to reach 1.5 million daily users. Kontraband will post links to advertiser videos for a fee, even writing an editorial plug in the site's prominent "cool stuff" section.

"If you engage people, they don't care if it's branded," said Spalding. "If it's good quality, people will pass it on."

For that reason, not all are convinced that viral efforts need outside help to goose their popularity. Daniel Stein, CEO of EVB-the digital agency that created OfficeMax's viral hit "Elf Yourself" site-said experience in campaigns is more important, giving his shop a set of best practices to follow. Christian Dietrich, head of the gaming unit at Tribal DDB-the agency that created viral sensation for Phillips-said smart creative that speaks to a target audience will win out, and a low-key approach is often better to achieve grassroots appeal. "There's too much risk of it coming across as false or commercial," he said.

But for all the efforts to bring a measure of predictability to viral campaigns, their nature ensures they can never truly be planned and executed like traditional ad campaigns, some agency executives assert. "No matter how much testing you do, there's no promise it'll be successful," said Stein.