Pure Viral Marketing – A Pipe Dream?

We hear it all the time: “We need a viral marketing strategy.” Or “we’re prepping a series of viral videos.” Or “we’ve got the best viral game in town.” Inspired by a series of best-selling business books with elegant but oversimplified narratives, our industry has been led to believe there is some magical, attainable viral force waiting to be unleashed. Viral marketing has become such an abused cliché that many people now just assume it to be a law of marketing — at face value. But it’s not.

Below is my MediaPost column this week, inspired by my recent meeting with Duncan Watts, professor of sociology at Columbia University, and an overdue challenger to many false assumptions associated with viral and influencer marketing. I didn’t interview him for this piece, but I report some of his conclusions published in a recent article he co-authored in Harvard Business Review. As always, be sure to check out the column’s discussion forum at MediaPost.

Pure Viral Marketing – A Pipe Dream?
August 17th, 2007 by Max Kalehoff

With traditional mass marketing losing effectiveness amid splintering audiences and rising costs, “viral marketing” has somehow seeped into our veins as an answer to the problem. Imagine that: a low-cost strategy whereby everyone embraces your insatiable commercial message as credible, and infects everyone they know. Then, those infected people infect everyone they know, and so on and so on. Hey, it’s almost free marketing! Who needs traditional anymore?

We hear it all the time: “We need a viral marketing strategy.” Or “we’re prepping a series of viral videos.” Or “we’ve got the best viral game in town.” Inspired by a series of best-selling business books with elegant but oversimplified narratives, our industry has been led to believe there is some magical, attainable viral force waiting to be unleashed. Viral marketing has become such an abused cliché that many people now just assume it to be a law of marketing.

But it’s not.

Duncan Watts, a Columbia University sociology professor, has been challenging these viral-marketing assumptions with a series of mathematical models, healthy skepticism and humble logic. I met Professor Watts at our offices a few weeks ago, and his arguments are compelling. The fact is, for any successful viral campaign, there are many, more that fail miserably.

In a recent byline article in Harvard Business Review, Watts explains the standard viral-marketing model, an analogy with the spread of infectious disease:

“It assumes that one starts with a seed of individuals who spread a message by infecting their friends, where the expected number of new infectious people generated by each existing one is called the “reproduction rate,” or R. When R is greater than 1, each person who gets the message will, on average, spread it to more than one additional person, who then does the same thing, and so on, leading to exponential growth in the number of people who receive it — an epidemic. By contrast, viral messages with an R of less than 1 are generally considered failures. That’s because purely viral campaigns, like disease outbreaks, typically start with a small number of seed cases and quickly burn themselves out unless their R exceeds the epidemic threshold, or tipping point, of 1.”

That’s right: like diseases, purely viral campaigns start with a smaller number of seeds and tend to quickly burn themselves out. Instead, Watts correctly notes that companies should adopt principles of viral marketing and mesh them with traditional mass marketing, a concept he calls “big-seed marketing.”

Watts explains:

“Companies, unlike diseases, can use standard advertising methods to create potentially enormous seeds. If the initial seed is big enough, then even if R is less than 1, the burnout process will persist for multiple generations, thereby, reaching many additional people. By providing social-sharing tools that are easy to use, moreover, marketers can reliably increase the reproduction rate of their message-an important point, as even small increases in R can dramatically increase the number of additional cases.”

For example, Watts suggested that if a campaign had a reproduction rate of 0.5, meaning that each generation is half the size of the previous one, then an initial seed of 10,000 people would pass it to 5,000 new recipients, and those 5,000 would pass it to a further 2,500. Eventually, the campaign would reach a total of 20,000 people — twice the number that would have been reached by the traditional campaign.

Watts’ proposition of big-seed marketing makes a lot of sense. Give up your pure viral-marketing pipe dreams and exploit the fact that campaigns can achieve reproduction rates that significantly build on media investment.

But here’s where I think we need more investigation. How do incentives influence reproduction rates, or pass along? Simple logic would suggest that many marketing campaigns have very low reproduction rates precisely because they are intrusive, irrelevant, coercive, annoying or lacking in credibility. As a result, advertisers tend to measure aided recall of their messages, not pass along – right?

But aside from factors that erode pass along, the success of any big-seed marketing initiative is also largely dependent upon the value that each generation of prospects believes they receive in exchange for a pass along — be it entertainment, prestige, exclusivity, monetary or other forms. The laws of big-seed marketing imply that word-of-mouth and alternative marketing techniques must tightly integrate with traditional media strategies. They are inextricably linked, but we need to better understand how.