|
Wednesday, April 30, 2003
More Meaningless Metrics?
The numbers reported by comScore Media Metrix's "new" search traffic measurement service (qSearch) seem to be the meeting with healthy scepticism from some observers (and, of course, comScore's competitors).
My issue with the numbers is that they seem to be bending over backwards to give Yahoo a favorable rating. I will leave you to speculate as to why they might do this. The claim that "Yahoo’s domestic leadership is driven primarily by the strength of its channel offerings, such as Yahoo! Finance and Yahoo! Yellow Pages" raises a red flag. Who told comScore to go and include those channels in their methodology? Is it 100% clear what is being measured? Is Yahoo actually getting too much credit as a search property for the normal activity that might be associated with content channels within a busy "portal"? Did comScore measure activity on Google's other channels, like Google Groups, Google News, Froogle, etc.? Perhaps even more importantly, to what extent are users willing to continue using Yahoo Search because it is powered by, well, Google?
Some other metrics seem contrived, as well. Before April 28, I had never heard of the "visitor-to-searcher conversion ratio." Sounds impressive, right? It puts Infospace's Dogpile and Metacrawler properties in a nice position of well-deserved leadership. And no doubt it is a stab at MSN, which enjoys a lot of visits from confused Internet Explorer users. But the fact of "converting" a visitor to a search engine site to a visitor who actually performs at least one keyword search is like calculating the ratio of high school students attending at least one class to students who enter the school on a given day. More meaningful info in that context would be whether the learning converts to high test scores. And in the e-commerce arena, at least from the search engine marketer's standpoint, anything short of knowing how many of those searchers convert to sales constitutes inconclusive data.
Oddly, then, small boutique "marketing laboratory" consultancies such as MarketingExperiments.com and Drilling Down seem to do a much better job of studying metrics with "teeth" than the established metrics agencies do.
Posted by
Andrew Goodman
View Posts by Category |
|
D'oh!
|
