Thursday, March 12, 2009
Since posting about what webcomics creators were seeing on GoogleTrends, I've received several chunks of detailed reporting from established webcomics producers agreeing with those who said GoogleTrends' measurements were off on the low side, and by a great degree. Even when compared with information in Google's own Google Analytics service, which, as Google says, it does not use in its GoogleTrends formula.
This is good to know (and heartening!), but it's also what makes the matter puzzling: If it is the case that GoogleTrends is tracking the traffic of this specific genre of websites more poorly than others, what's the reason? It would seem to be a problem worth exploring. I know from my days working in interactive media at a publisher that rankings from sites like GoogleTrends and Alexa are generally not used in advertising decisions — there are far more precise, auditable measures that publishers can provide. But if there's something peculiar to the way Google interacts with webcomics sites, it would be good to know about.
I have a few theories, but I suspect there are readers out there who have more facility with the nuts and bolts. Webmasters spend a lot of effort optimizing their sites for search engines; there are probably tricks to optimize for external auditing systems, as well. (Presuming we knew how the systems were doing what they're doing — which, in Google's case, we really don't...)