One of the great things about search advertising is that there is a wealth of hard data out there that can tell you how many people are searching for a particular keyword before you bid on it. When you serve advertising on a search engine results page you reach people who are searching in order to take an action either now or later.
On the other hand, advertising on publishing sites more closely resembles other forms of traditional print and television and share with those mediums a degree of uncertainty over whether the people seeing your ad are actually intereste din your message. The uncertainty grows when dealing with ad networks which can claim, as one former ad network employee tells me, to “reach 90 percent of the Internet” — an astounding and meaningless figure.
This is not to say that all ad networks are bad or that there is no value in the segment. At a time when many online publishers are finding themselves without inventory, ad networks have been a place to turn to for help. In fact while ad revenue for publishers has grown 32% year over year, revenue for ad nets was up 50%.
From a marketer’s prospective how do I know what I’m really getting from an ad net when they toss out stuff like a 90% reach? Apparently someone’s been scratching their heads at comScore over this very same issue because they have announced a new set of metrics for their August data that measures both “Potential Reach” and “Actual Reach.” I reckon that the actual reach figure is the one most likely to elicit a “Ruh-Roh” from the ad nets. Still there is a caveat – can you spot it in this excerpt from comScore’s press release?
“The “potential reach” measure will be a calculation of the unduplicated visitors to all sites with which each ad network has contracted to deliver advertising. The definitions will be based on written documentation and confirmation provided by the networks. This measure represents the largest potential reach a network could deliver if all parts of the network are used to deliver ads.
The “actual reach” measure will represent the reach of the ads that are actually served by the network during a given reporting period. In order to report “actual reach,” networks must participate by providing comScore with the appropriate identification protocols for rendered ads. Consequently, only participating networks will be included in the “actual reach” report.”
Give yourself a gold star if you zeroed in on the fact that the ad network has to participate and share data in order to deliver “actual reach” data. This will give a handy guide for the marketer as to which ad networks one would want to use. ComScore is a very reputable data source and a network that refuses to participate ought to be looked at with a wary eye indeed.
I’ll never forget a few months ago when a company came in proudly boasting that they had a %91 reach with their latest acquisition, beating Ad.com by 1-2 percentage points. With the majority of large Ad networks chasing each other for percentage points and no on really offering a clear differential, it should be interesting to see who is really willing to take the risk to show their actual numbers. It may prove to be a way to reboot competition and by showing realistic numbers, a couple of percentage points might actually mean something again.