I'm not sure more is better when it comes to creating customer segmentation analysis based on social media activities and attitudes. Still, the virtual discussion is fruitful, and it makes me think about demographics vs. psychographics vs. generational schemes. And to wonder how much the models we create now will have to evolve in the future.

I'm a little late to Gartner's social marketing party, but (new) colleague Jeremiah Owyang tweaked me to it. Gartner analyst Adam Sarner goes into a little more detail in this Forbes column.

Gartner calls its system a generation, but disavows age and other demographics. (And predicts more marketing spending online than off- in ten years.) Whatever. Jupiter, Forrester and most marketers and programmers I know still think a lot about demographics. Still, I'm partial to a more "psycho" approach -- I've been convinced over the years that attitudes and experiences are highly predictive of future behavior. And I've been working on Millennials analysis as well.

But the coolest bit of the Gartner approach, at least as it's laid out in the column, is the idea that since people have multiple personae online, marketers and programmers need to match up with those, rather than with the individual behind them.

I wonder if social media behavior aligns more with generation, with brands, or with activities. Do you update your Facebook status, but never post reviews? Or do you review movies on Flixster but not on Amazon? Or search for books the same way you search for movies, let alone healthcare products?

Modal, contextual personae. Now that's a segmentation that'll generate consulting $$ and agency fees for years to come....