Mismeasuring Influence?

The algorithm missed this one

Lee Iacocca once remarked that he wished he had learned earlier in life how important the ability to communicate is. That’s a sentiment that I encounter frequently among my executive communications coaching clients, especially those who rose through the ranks on the strength of their strong technical skills but reach a level where they find that their effectiveness now rests on their ability to get things done through others.

But you don’t have to rub shoulders with top management to benefit from the ability to influence others, a fact brought home by a recent article in the Wall Street Journal. The article says, “Armed with reams of new data, companies including giants such as Procter and Gamble Co. and Cisco Systems Inc. are seeking out ‘influencers,’ or those who are particularly well-connected and trusted by their peers.”

When they find them, they use them to help spread information, deal with change, and drive important projects, and the influencers receive more formal recognition and reward as a result.

So far, so good. It certainly aligns with the theme of this blog that persuasive communication skills are critical to personal success. So why was I bothered by reading the article?

Because I know how easy it is for companies to get carried away with the latest and greatest trend, and to base decisions on bad data just because it’s easy to collect.

How do they measure informal influence? The new data includes mining email or Salesforce.com Chatter to map webs of communication and identify top influencers. Salesforce.com uses “algorithms”[1] to analyze who has the most followers and most posts and so arrive at individual influence ranks, and invites the top 20 to their global senior management meeting.

Maybe those algorithms are telling them something useful, and maybe not. As any regular user of Facebook knows, there are some people who post stuff constantly, but my own personal algorithm (i.e. SWAG) tells me there is no correlation between how often they post and how influential they are. In fact, if I were to guess at a correlation, I would say that the people who are actually getting things done have less time to spend on their Chatter score.

Second, when you start rewarding people for their influence scores, do you think some of them might just game the system? As the quarter end approaches, do you focus on getting your project done, or do you spend time generating posts to gin up your score?

Third, good managers already know who the top influencers are. They know because they are walking around, talking, and listening—not because they are pulling up a score on their computers.

Certainly, influence is one of the most valuable personal assets you can cultivate within an organization, but don’t ever kid yourself that it can be developed and measured by social media algorithms. The best medium for influence is still an eye to eye conversation.