Influence Is Not Star Juice

"Influentials" is a funny word that makes me think of someone sick with influenza, with a runny nose and a feverish delirium.  That aside, Merriam-Webster offers an amusing definition of influence as "an ethereal fluid held to flow from the stars and to affect the actions of humans." Which is exactly how our industry imagines the "influentials" -- the more star juice they have, the more influential they are. (Here's an example from Brian Solis: "Influence is the ability to cause desirable and measurable actions and outcomes.")

Only it's not like you can buy star juice by the gallon, and so nobody can quite figure out what "more" means.  Klout's kscore measures observable outcomes: clicks, retweets and other similar stuff. Other formulas are fancier: "Influence = (Personal Brand * Knowledge * Trust2)". These and similar approaches to measuring influence are wrong for two reasons.

One is that considering only observable behavioral outcomes such as clicks could lead us to confuse influence with conformity, authority and power:
"Conformity occurs when an individual expresses a particular opinion or behavior in order to fit in to a given situation or to meet the expectations of a given other. Power is the ability to force or coerce someone to behave in a particular way by controlling her outcomes. Authority is power that is believed to be legitimate (rather than coercive) by those who are subjected to it." (source: pdf)
Social influence is a social phenomenon (no kidding), doesn't exist in a vacuum, and involves at least two people actors* -- an object A and a subject B. And the bigger problem with the "star juice" definitions is that they focus exclusively on A, count the number of Bs, and completely ignore the relationship between the two.  In other words, these definitions suggest that influence is an absolute attribute of A, sort of like weight, only expressed in the number of Bs.

Herbert Kelman, a long-time scholar of social influence at Harvard, defines (pdf, url) influence as an outcome of interaction between A and B:
- "Social influence can be said to have occurred whenever a person changes his behavior as a result of induction by another person or group.
- "The definition of social influence implies at least some degree of resistance to change that has to be overcome."
- "Social influence represents an aspect of the relationship between [A] and [B] within a social system in which both occupied specified positions."
- Influence can be positive or negative. "Negative influence refers to a change in a direction opposite to that induced by the influencing agent."

To sum up:
- "Influence" describes A as much as it describes B
-  Influence is situational
-  Not all behavioral change is an outcome of an "influence situation"
-  On the other hand, a lack of change can also be an outcome of an "influence situation" (negative influence)

In this context, measuring someone's influence with a yardstick makes about as much sense as reading the distance from Boston to New York from a thermometer, but at least you'll get some number.

(...more soon)

*Correction: A & B do not have to be individuals, either of them (or both) can be a group.


  1. I don't follow your final summary point: 'On the other hand, a lack of change can also be an outcome of an "influence situation" (negative influence).' In Herbert Kelman's paper social influence refers "to socially induced behavior change." If there is not a change, then his argument seems to be that the interaction is not an "influence situation." My understanding is that negative influence still results in a change, but it is not a change in the direction that object A was trying for.

    Thank you for writing the article, it is an interesting topic.

  2. I agree with your assertion that social influence is an interaction of two entities rather than one acting on another. With that said it’s interesting that Merriam Webster also defines influence as "the act or power of producing an effect without apparent exertion of force or direct exercise of command." This definition implies exactly the opposite--influence occurs without force; the effect simply happens without a true interaction between A and B. This sounds a bit like the "ethereal fluid" that "affects the actions of humans" that it also mentions. You mentioned that these definitions could create confusion and I agree that the “conformity” definition does work here. This definition states that the entity A does not exert a force over B. Instead, societal pressures act as the force that changes the opinion of B.

    In the advertising world the definition of social influence is extremely important to understand but also very difficult to exactly define. I think that your summary offers a useful base for the definition of this concept.

  3. @Shannon - The key in your definition is that influence is "an act of power", and not the power itself, and people seem to confuse the two.

    @Doug - I am afraid I haven't expressed it clearly, and I agree with you. Negative change is still an outcome of an "influence situation".

    The last paragraph refers to the entire article, not just the piece about negative change. What I meant is that measuring influence as an absolute attribute of an individual is meaningless because influence is situational and refers to both actors, not only A.

  4. Ilya,

    Good stuff. I like to think of relationships as having a kind of "currency", which of course means value, but also a sense of "now" value: how much this value has in the current context of A and B's situation, exchange, and/or relationship.

    Thanks for writing, very thought-provoking.

  5. Social influence is a social phenomenon (no kidding), doesn't exist in a vacuum - Totally agree .

    Its like a game played well assertively and positively.

    The last decade was about connections which was enabled by the Facebook Graph whose construction has reached a peak and a saturation point.

    The next decade will have a focus on the influencing behavior for which the construction does not have any set foundations and the work has just begun.

    Read More:

  6. Not all behavioral change is an outcome of an "influence situation" - Precisely. Thank you for sharing these thoughts.


I am moderating all comments to weed out spam (there's a lot of it). Comments are usually approved within a day.

Related Posts with Thumbnails