AD NETWORKS: IN FOCUS
Published: January 22, 2007
What Smart Buyers Want from Ad Networks
 
Slice inventory thin

The sheer number of affiliates within a network can be mind-boggling to an agency buyer. It's important for networks to have a broad array of targeting options with which to carve up the billions of impressions an ad network might serve on a monthly basis. Agencies often win clients in part based on how they're able to whittle down wide swaths of inventory to give clients the most targeted inventory possible. Ad networks should prepare their arsenal of targeting filters to be sure they can conform to advertiser requests in this regard.

Adding the most popular targeting options like geographic and demographic targeting is pretty simple, but buyers are increasingly using complex targeting options on networks, like behavioral and contextual targeting and domain-level targeting.

In integrated media campaigns, online media often play the role of covering the sweet spot of a target audience, zeroing in on the prospects most likely to convert. Many planners will cover the low-hanging fruit before they move on to less stringent targeting requirements. Networks should thus be sure they can accommodate all or most of the targeting methodologies in demand by buyers.

It almost goes without saying that the technological kinks should be worked out of the network ad serving systems before buyers stake their reputations on placing a buy with a network. While a lot of networks claim they can target by a wide variety of criteria, buyers often hit snags during implementation.

For instance, an ad network that claims to be able to target by psychographic profile might have only a handful of profiles in its system as compared to its total available inventory. This can cut down on the number of ads that can actually be served to those user profiles over the life of a campaign. The more of an opportunity that a network has to identify behavior and gather data, the more success they’ll have in building up their base of profiles and the more targeted inventory they’ll be able to offer. This is one of the reasons behavioral networks like TACODA are so compelling.

It is rather difficult to forecast inventory in such situations. Network reps should err on the side of caution and lowball estimates to make sure they're able to serve what they claim they can serve.

« Previous page | Next page »