• by Dave Morgan , Featured Contributor, November 3, 2011

TV advertising needs frequency capping. The recent availability of large amounts of anonymous set-top-box viewing data for tens of millions of U.S. viewers from companies like Kantar, TiVo, TRA Global and Rentrak has make TV advertising extremely measurable, particularly when it comes understanding the specific distribution of ads to each viewing household. TV ads may not yet be addressable like they are on the Web, but they are becoming similarly measurable in many respects, particularly when it comes to actual audience reach and frequency.

I recently spent some time digging into the ad viewing data that my company has collected, and was struck by the imbalance in the frequency distribution of ads in most mass awareness national brand campaigns. Surprisingly (to me, at least), the campaigns for most large reach advertisers tend to deliver an enormous proportion of total ad impressions to a very small amount of the total audience they reach, and very few ads on average to the majority of the total ads they reach.

In the movie category, for example, the campaigns for most wide theatrical releases, which might have a total reach of 75-80% of all TV viewers over the three weeks preceding the film’s release, tend to deliver a full two-thirds of their impressions to only 20% of the total TV audience. The next 20% tend to get another 20%-25% of the ads, meaning that 85%-90% of all campaign ad impressions are only going to 40% of the TV audience.

This must be because those folks have been targeted as “frequent movie goers”? Actually, that’s not the case. When you cross this ad viewing data with Nielsen’s flags for movie-going habits of those viewers, you find that those 40% are just as likely to be “Not Frequent Movie Goers” as they are to be “Frequent Movie Goers.” More than one-half of Frequent Movie Goers watching television tend to see only one or two ads in any theatrical release campaign, while the other half receives an average of eight to ten ads in that time period.

What is unique about the half of viewers that received so many ads? They tend to be heavy viewers of TV generally, and heavy viewers of broadcast networks in prime time specifically. Many of them see more than 50 ads in each movie campaign, and some more than 100. I don’t know how many ads it takes to make someone aware of a movie opening and give them a sense of whether or not they might like it, but it’s got to be less than 50 and more than none (what 25% of the viewers received).

In my previous life in online advertising, we had a pretty simple answer for that kind of problem: frequency capping. You used cookies to limit the number of times any particularly browser might see a particular ad on a daily, weekly or campaign basis. While it was far from foolproof, it did have the desired effect of reducing the amount of waste in a campaign by limiting the number of ads delivered to the heaviest users of certain sites. TV’s equivalent? Using highly granular direct-measured viewing data to limit the number of ads shown to shows with redundant audiences, and redistributing those spots to shows adding incremental reach to the campaigns (maybe long-tail networks).

Is this likely to happen soon? Certainly not overnight. TV media planning processes change very slowly. However, with the data now available, and with CFOs and procurement officers pushing marketers and their agencies harder and harder to do more with less, it’s only a matter of time. What do you think? Will we see some form of “frequency capping” on TV?