Behavioral ad targeting has been getting a lot of attention in the press lately, especially around the US Congress’ interest ing the technology. In general ad targeting of various shapes and forms is also becoming a busy space for startups - various types of targeting technologies trying to understand the users intent, and provide them with an appropriate advertisement.
When I looked a bit closer into what most of these ad targeting companies do, it turns out that after they have used whatever mechanism (contextual, behavioral, demographic, psychographic etc.) to decide who you are and what interests you, they translate this into one of a very small number of consumer segments, pick an ad for that segment and display it to you. So all that fancy computation upfront to provide a canned ad. Seems kinda of a waste. Wouldn’t it make more sense to create a personal, data driven ad (from one or more advertisers) to leverage that information? That would be the “holy-grail” of true 1-1 personalized advertising.
This growing impedance mismatch between the fine grain targeting ability of ad networks vs. the coarse grain delivery capability of advertisers is going “short-circuit” the ability of these targeting technologies to show their full potential.