Share this content

Influencer marketing: Effective or defective? (Part one)

8th Aug 2008
Share this content

The concept of aiming marketing communications at a small number of highly influential networkers has grown in popularity since the advent of Web 2.0. But the validity of influencer marketing is now under attack. In the first of a two-part report, Neil Davey explores the issues that surround influencer marketing.


By Neil Davey, editor

With the media landscape in a fragmented state, and the Web 2.0 revolution encouraging rising numbers of consumers to make purchasing decisions based on word of mouth, the concept of influencer marketing has never been so appealing. And it is a concept that has been around for some time.

The idea was proposed as far back as the 50s, when sociologists Elihu Katz and Paul Lazarsfeld spoke of 'opinion leaders' who catalyse trends in their book 'Personal Influence'. But the concept acquired wider appeal in recent years as first Malcolm Gladwell discussed the phenomenon in 'The Tipping Point' and then marketing experts Ed Keller and Jon Berry hailed its value in 'The Influentials'. In the last few years, the topic of influentials has itself become the talk of the town.

"Influencers are so disproportionately represented online that it is a great way to connect with the influencers who then will spread the word through every available channel to others."

Brad Fay, Keller Fay Group

"The reason why brand owners are so interested in them is that if you can tap into these influencers then hopefully you can shape their views so that they can influence others in a way that suits your marketing objectives," says Alex Bollen, research director of Ipsos MORI, which has been studying the behaviour of influentials in the political arena – what it termed 'socio-political activists' – for decades. "That is also set against a backdrop of fragmentation of the media landscape which has meant that it is much harder to connect with people. Firms are looking for new ways to cut through the huge mass of advertising messages and if you can tap into an influencer network then that is an incredibly effective way of doing it. But it is a big 'if'."

The Web 2.0 revolution has only served to further ramp up interest in influentials, as the importance of opinions and recommendations in consumer decisions has increased as social networking has proliferated. Reflecting this, research by Ipsos MORI highlights that the proportion of consumers identifying word of mouth as their best source of ideas and information has increased from 67% in 1977 to 92% in 2005. But the advent of Web 2.0 has had other implications for marketers according to Brad Fay, co-founder of the Keller Fay Group and a co-chair of the Word of Mouth Marketing Association's (WOMMA) Influencer Council.

"The internet attracts people who are influencers," he explains. "Our view is that even though a great deal of conversation and recommending happens offline, influencers are so disproportionately represented online that it is a great way to connect with the influencers who then will spread the word through every available channel to others."

"In the last couple of years the concept of influentials has really bubbled up to the top," agrees Steven Hershberger, principal of ComBlu and a co-chair of the WOMMA Influencer Council. "Branding agencies have realised that these individuals are to some degree the nucleus of social interaction, sometimes the well of information (whether anecdotal or technical) and sometimes the catalyst."

Misleading representations of reality

But is this bubble about to burst? Last year, Duncan Watts, a network scientist at Columbia University, launched the first significant attack on the influentials hypothesis. Watts' suggestion that the hypothesis was neither supported by empirical evidence, nor in sync with accepted theories of interpersonal influence, made the headlines and rocked the marketing world. Watts pointed to research that he had undertaken with his colleague Peter Dodds that demonstrated through the use of simulation models that under many circumstances influentials would at best have a modest impact on public opinion change relative to other individuals. He concluded that when word of mouth was successful, it was more related to the structural properties of the entire social network than the influence of a small number of special individuals.

"Organisations run a campaign and if it seems successful, they assume that it must have reached the influentials; and conversely, if it doesn't succeed, they assume that it didn't. But they don't actually try to identify these influentials ex-ante, or design a dedicated word of mouth campaign around targeting these individuals."

Duncan Watts

Whilst emphasising that his work did not undermine the relevance of interpersonal influence or the efficacy of word of mouth marketing per se, he proposed that an appreciation for the true complexity of influence would be beneficial. For instance, organisations could devote more time to detecting and fostering success that is already happening (whether anticipated or not), rather than finding who should be targeted for a product to succeed. Alternatively, he suggested, traditional mass media techniques, such as mailing lists, might benefit word of mouth marketing, if they were supplemented with viral functionality.

Watts did concede that with the emergence of Web 2.0, advances in online technologies such as the tracking of blog postings did offer some promise for more objective observational models. But even then, he warned, challenges would still exist in terms of identifying who influenced whom rather than merely who communicated with whom. In his paper 'Challenging the Influentials Hypothesis' he concluded that until more sophisticated methods for observing interpersonal communication exist that integrate other potential sources of influence and subsequent consumer attitudes, "identifying influentials will be at best crude approximations and at worst confusing and misleading representations of reality".

But Watts believes that a more damning indictment of influencer marketing has been delivered subsequent to the publication of his findings. "I have since had the opportunity to talk to quite a few marketers about influentials and how they think about them," he says. "And what has surprised me about these conversations is that as much as people talk about the importance of influentials, they don't typically implement marketing strategies that explicitly try to identify or target influentials. There are a few firms that do claim to implement such approaches (although what they actually do is usually not terribly clear), and some of the large advertisers devote some attention to finding and recruiting people they think will become brand advocates, etc.; but my impression is that all this effort is small potatoes relative to overall marketing budgets.

"Mostly when I ask large advertisers how they target influentials, they answer that they don't - at least not directly. Rather what they do is conduct a whole range of advertising strategies - some traditional, like TV ads, and others, like 'viral' or 'guerilla' campaigns, that are relatively novel - and then sort of hope that the influentials will pick up their message and generate a big multiplier effect. So if they run some campaign, and it seems successful, they assume that it must have reached the influentials; and conversely, if it doesn't succeed, they assume that it didn't. But they don't actually try to identify these influentials ex-ante, or design a dedicated word of mouth campaign around targeting these individuals.

"What's ironic about this is that it isn't actually so different from what we concluded based on our theoretical models - that is, targeting influentials is probably not worth the effort, so simply target people you think will be receptive to your message, and hope that they pass it along to other receptive people. In a largely random world, success is hard to predict, so try lots of different things, and don't put all your eggs in one basket."

So, what does the marketing community make of these criticisms - and is there any truth in them?

Read part two to find out.


Replies (2)

Please login or register to join the discussion.

By jeffgordon
18th Aug 2008 22:49

Woven into the overall academic approach to this commentary on Influencer Marketing is a fundamental truth: That is that most marketers do not understand what this type of marketing is all about. The truth is, that influencer marketing has more to do with intimate selling techniques than it does mass marketing techniques. Colleges and MBA programs crank out marketing people all day long...but sales ability is something no school can teach you.

My perception of this article is that the author is questioning the validity of influencer marketing based on the fact that most polled "marketers" don't fully understand how to isolate an "influencer" or measure what effectiveness they have on branding or bottom line sales. So the article should be about the shortcomings of most marketers.

"Influencer Marketing" is akin to "leverage selling" and is no different in concept than having a "spokesperson". Industry has been doing this for a long time. But the process to isolate, secure and nurture influencer teams is very foreign to the average marketing person. I suspect that many major companies are moving to a Influencer Marketing/PR model and dedicating poorly equipped field marketing professionals with no understanding of the salesmanship that is required to execute an effective IM campaign.

Thanks (0)
By AnonymousUser
19th Aug 2008 09:42

Thanks for your comments Jeffrey.

Certainly I intended the article to reflect the fact that there are concerns over issues relating to influencer marketing such as the identification of influencers in the first place. But this is really only part of the story.

Influencer marketing is really very different from merely having a spokesperson, and one of the biggest debates rages over the social dynamics that are involved in the process of 'influencing' others. Can they be predicted and leveraged? Or are they fundamentally random?

Ultimately, the article was designed to highlight that irrespective of these concerns, influencer marketing is rapidly growing in popularity - and that firms should stop to consider the issues before just joining the herd ( does this behaviour in itself make a case for the existence of tipping points and memes ;-) )

I hope the second part of this article in particular emphasises this.

Thanks for your interest,

Neil Davey

Thanks (0)