Today I saw 2 more threads in the ongoing debate over whether social media popularity is a good way to measure influence.

First, my colleague Duncan Brown writes that Google is launching an AdWords-style SEM program across big social networks.

As an online publisher, I can see how this Google program makes perfect sense for media buyers. It will play from Madison Avenue to Main Street. After all, the big advertisers say they plan to shift their remaining 2008 and 2009 spending, cutting traditional ad spending while increasing spending on word of mouth and other forms of social media. (For the latest CMO study visit Epsilon; hat tip to Ken Rutowski for flagging it in his newsletter.) Google is offering just the right media product to pick up those extra dollars and euros. I’ve got no issue there.

However, I do see a potential downside. Call it collateral damage. Google is portraying the program as a measure of influence. Duncan describes the confusion this could cause:

“If Google’s plans get more firms to talk about influence, then fine. But I fear that it will dumb influence down to a few ‘magic’ numbers that have tenuous relevance to real influence.”

Meanwhile, Graham Hill and I compared notes this morning on Peter Kim’s post, “Influencer Lists as Ego Traps“. We came up agreeing, in Graham’s words:

“Popular people are not necessarily good influencers. And influencers are not necessarily popular. There is much more to it than that.”

We’ve got some very bright people on both sides of the debate — those advocating that we equate influence with popularity/connectedness, those advising against it. Neither side is ready to blink.

In the end, the media buyers may have the final vote on whether online popularity is the path to the influentials.