In their efforts to woo traditional advertisers over to interactive, some are advocating for the banishment of the impression. Tom argues that the much-maligned metric has its uses.
Jim Meskauskas brought up some interesting points in his column this week about how media is bought in the online realm as compared to the offline world. He is quick to point out that when online media buyers purchase activity on a cost-per-thousand (CPM) basis, most times it is on what he terms a "gross CPM" or an untargeted CPM, which doesn't jibe with how media is bought in the traditional world.
True indeed. We are working with two incompatible metrics here. But does that mean we should kill the concept of the impression in online media? I'd argue that we shouldn't.
There's been a lot of discussion about the audience-based pricing model in online media and about what action the industry needs to take to make it work. Interjected in these discussions have been mentions of the surround session model introduced by the New York Times. Though surround sessions do draw us closer to an audience-based model, I think that some people who have been participating in these discussions may have become confused about what would allow the industry to measure its interactive ad activity alongside its traditional activity.
From my traditional days, I remember that all media buys could be backed out to a measure of gross impressions. That is, reach and frequency against a given audience could be converted to targeted rating points (TRPs) against that audience. The number of TRPs would depend on the number of targeted impressions versus the audience size. For instance, if there are 20 million U.S. males ages 18 to 24 and I buy 40 million targeted impressions against that audience, that would represent 200 TRPs (40,000,000/20,000,000*100).
TRPs do not account for duplication, so 1,000 TRPs against women ages 18 to 54 does not mean that everyone in that universe has been reached with a frequency of 10. Some people within that audience may have seen an ad 20 times, while some may not have seen an ad at all. If you took a measure of all your TRPs, divided by 100, and multiplied by the size of your audience, you would end up with a measure of gross impressions against that audience.
This begs the question: Why can't we have TRPs in interactive media? If sites can target their ads to specific demographic audiences, why can't we convert those gross impressions to TRPs? Personally, I don't see any reason why we can't.
To convert gross impressions to measures of reach and frequency in my traditional days, I used to count up all my gross impressions by medium. I'd then insert these figures into an application called MixMaster, which used the different reach and frequency figures I would input for each medium to develop an aggregate reach and frequency based on the media used to reach the target.
Why wouldn't this work to help interactive media measurement in the context of an integrated communications plan?
Maybe what we really need is a measure of TRPs in interactive so that the folks who want to measure interactive the same way they measure print, radio, TV, and outdoor can do so. But this doesn't mean we need to kill the concept of the impression. Rather, we need to refine that concept such that impressions become targeted impressions against a given audience.
It seems to me that what we really need to do is move to TRPs and develop reach and frequency curves for online media. These curves would have to take into consideration how many sites are used and the size of those sites (unique users versus total traffic).
Although in some respects, using demographic targeting does sell short some of the capabilities of interactive media, it still has an advantage over traditional media, in that advertisers can buy only the targeted impressions they want, rather than buying a page in a magazine and knowing that a certain percentage of the people who see it will be outside the demographic target. Thus, a site such as the New York Times can segment a home page placement in such a way that ads from GM appear when men visit the home page, and ads from Procter & Gamble are served when women visit. Two advertisers occupy the same theoretical "space," each pays less than it would for buying a persistent placement, and they get only the prospects they want.
I wouldn't kill the impression just yet. It looks like it could still be the base currency of our industry... with just a little refinement.
Tom Hespos heads up the interactive media department at Mezzina Brown & Partners. He has been involved in online media buying since the commercial explosion of the Web and has worked at such firms as Young & Rubicam, K2 Design, NOVO Interactive/Blue Marble ACG, and his own independent consulting practice, Underscore Inc. For more information, please visit the Mezzina Brown Web site. He can be reached at firstname.lastname@example.org.
2015 Holiday Email Guide
The holidays are just around the corner. Download this whitepaper to find out how to create successful holiday email campaigns that drive engagement and revenue.
Three Ways to Make Your Big Data More Valuable
Big data holds a lot of promise for marketers, but are marketers ready to make the most of it to drive better business decisions and improve ROI? This study looks at the hidden challenges modern marketers face when trying to put big data to use.
December 2, 2015
1pm ET/ 10am PT
Wednesday, December 9, 2015
5pm HKT / 5am ET