Thanks to the continuing evolution of social media, there’s a lot of talk at the moment around influence, identifying and understanding it in action, as well as how this can be used commercially. 

Just last week my colleague, Matt, wrote an insightful post about the intricacies and accuracy of this, but recent conversations I’ve had with quite a few people makes me think that on a wider basis, the fundamentals around this increasingly complex area are misunderstood, and in some instances, overlooked altogether. 

I’m a firm believer that in any business discipline, the basic framework on which the subject sits needs to be identified before any aesthetic layers are added.

So, with this in mind, I want to explore the core concepts of social influence and how this can be looped back into marketing theory. Let’s strip it back to the bare bones… 

Social influence

Influence as a social and psychological concept is quite complex, but has been broken down into three core forms: 


Where individuals publicly agree with others, but keep their personal opinions private (even if they disagree in reality) in order to gain a satisfaction from the effect of this within their social circle. 


Individuals are influenced by someone who is known publicly – such as celebrities – and demonstrate a want to be associated with them, often by trying to identify themselves with the person in question.

This is arguably the most widely recognised influence form, given that advertisers have been using celebrity endorsements for decades. 


Individuals agree both publicly and privately, based upon their own personal beliefs, thoughts and opinions. There is no conflict of thought vs. action. 

Social impact theory

Sitting on top of the tripod framework of social influence is another three-pronged layer: social impact. Loosely, this considers the core elements that are key to generating individual responses to social influence: 


The personal importance of the influencing individual or group to the person being influenced. This is especially applicable to the “Identification” element of social influence. 


The closeness – physically and emotionally – of the influencing individual or group upon the person being influenced, specifically at the moment of persuasion, rather than a longer-term concept. 


The number of influencing individuals, or of those being influenced, can affect social impact. 

Complexities of influence 

So, great: the basis is there, and it’s relatively easy to see these theoretical elements as a reality inside the social media landscape.  

However, building upon this framework, it becomes slightly more difficult, as there are numerous interpretations of influence itself and, of course, social influence within this. 

There is a lot of academic thought in to this area, with a great deal of it bordering on mathematical equations that feel akin to quantum mechanics.

Most of the time, the outcome is the same: An individual “with privileged access to information can influence the decision of [their] peers in two fundamentally different ways. First, with [their] decision to hide or transmit [their] private information. Second, with [their] choice to acquire or forego extra evidence”. 

But what does this actually mean in practice? Well, it loops neatly back to the combined layers of how people influence others and to what extent this influence has, based on multiple factors. 

This of course, throws everything into chaos. Given the existing theories that are born from this relatively simple framework, such as cognitive dissonance, elaboration and reasoned-action models, trying to accurately focus on the core factors that are needed to identify social influence which can be used for marketing purposes becomes extremely difficult. 

However, the extremely smart Geoff Livingston recently wrote an excellent piece about the state of social influencer theory, in which he identifies the landscape of influence as it currently exists in a commercial sense.

Together, with Jess3, he’s built a framework that simplistically identifies the relationship between all the different existing influence models. 

But what are the current models? 

Livingston’s article, which also condenses a large discussion from his excellent book, Welcome to the Fifth Estate: How to Create and Sustain a Winning Social Media Strategy, goes on to succinctly describe the development and evolution of these influence theories.

Unfortunately for me, he’s done such a good job of explaining them, I’m forced to quote him pretty much directly: 

The Tipping Point 

Malcolm Gladwell (2000) Movements are caused by three types of influencers: connectors, mavens (subject-matter experts) and salesmen. Examples: Old Spice Guy, Dell Listens.

Six Degrees/Weak Ties 

Duncan Watts (2003) Data analysis shows influencers rarely start contagious movements; instead, average citizens provide the spark. Examples: Egyptian revolution, Tumblr – Digg events.

One Percenters 

Jackie Huba and Ben McConnell (2006) It is the content creators amongst internet communities that drive online conversations. Examples: Lady Gaga, Ford Fiesta.

The Magic Middle 

David Sifry (2006) The middle tier of content creators and voices break stories, and discussing that trickles up into widespread contagious events. Examples: 2008 Obama election, Motrin Moms.

The Groundswell 

Charlene Li and Josh Bernoff (2008) Movements start within communities, and leaders rise up out of the community and can have many roles including content creator, critic and collector.

Examples: Haiti earthquake texting, Pepsi Refresh Project.

Trust Agents 

Chris Brogan and Julien Smith (2009) Influencers are people who build online trust and relationships with communities that look to them for advice and direction.

Examples: Gary Vaynerchuk’s Wine Library TV, Republican Party’s #FirePelosi campaign.

Free Agents 

Beth Kanter and Allison Fine (2010) These trusted influencers are independent of traditional command and control organizations and crash into walls of storied culture.

Examples: @BPGlobalPR, Robert Scoble at Microsoft’s Channel 8.


(2010-11) Influence can be quantified by online actions taken by a person’s community, including retweets, mentions, comments and more. Examples: Klout, Empire Avenue.

What does this mean? 

At the most basic level, it’s suggested that all these theories fall (loosely) into two separate distinctions, based upon the earliest influence identifications from Gladwell and Watts.

However, what I find interesting is the evolution of influencer theory when charted against the development of the social web. 

I’d even go so far as to argue that there is no fixed, silver-bullet model of influence: it’s massively dependent on context. As Watts himself acknowledges, “marketers have been chasing influencers for a decade and they haven’t found them”.

That said, in my opinion, Watts then goes on to absolutely nail the crux of the issue:

If you actually want to figure out how influence works and you want to intervene to exploit social influence, you need to start doing it in a more scientific manner, which means you need to do experiments and collect data.

Given the sheer volume of social data and the methods that currently exist to assimilate this, (something discussed in Econsultancy’s recent Internet Marketing Strategy Briefing), I think it’s possible to identify influence – and influencers – along with the fact that no model-fits-all exists. It’s dependent on understanding your social data and a clearly defined objective, alongside a carefully structured strategy built around these factors. 

This is why specific organisations, campaigns and individuals are successful, they focus on the social information available and consider which social influence model is the best approach to take.

Quanitifiable metrics can play a key part in this, and isolating the “magic middle” (those with disproportionate influence) is important. Obviously, there are rules to the exception, where uncontrollable or outside factors can play a part in influencing people.

Again, the makes the issue a bit more elaborate, so what was a simple framework of different, but connected models, becomes expanded: 

Inside this, the elements of influence can be used. I’m quite a fan of Robert Cialdini, who has established six “weapons of influence”: 

  • Reciprocity.
  • Commitment and consistency.
  • Social proof.
  • Authority.
  • Liking.
  • Scarcity.

These fit tidily into the three parts of social impact theory (at the beginning of this post), but again indicate that there is no single model or factor of influence that can be applied to all situations – and again highlights the increasing complexity of the matter.

Marketers need to move beyond looking at single influencing metrics, ideas and the successes (or failures) of others, given that there are interwoven, paradoxical elements involved in the issue. 

This is something that I could wax lyrical about for far too long, but I feel that is actually best summarised by Livingston, who suggests that:

While we try to force finite concepts onto this [influence] zeitgeist, we are at the same time evolving the way it works, changing our own information patterns as we seek to understand, evolve and expand this.