Until the Web, media measurement has been more artistic statistic than scientific exercise. Still, finding ways to express interactive advertising's impact by using traditional advertising's nomenclature and units of measure is in vogue. For good reason.
Not that there has been a lot of talk about it within ClickZ lately, save for a few articles, but looking at online advertising through the lens of traditional advertising is becoming more and more the vogue.
Last summer I wrote about the need to alter the media currency in the online advertising space, moving the market from an exchange of impressions to that of audience or circulation, the way it is done in print (see “An Alternative to Impressions?“).
More recently, talk around the campfire turned to representing online advertising in the form of gross rating points (GRPs), the way it is done in broadcast.
Since then, there seems to be more interest in finding ways to express interactive advertising’s impact by using traditional advertising’s nomenclature and units of measure.
This is not a bad thing. It helps to translate some of how the Web might work for advertisers into terms that traditional marketers can understand and into forms they can look at side by side with the rest of their advertising activities — and so understand their goals, objectives, and impact as one whole.
Certainly, some who have grown up in advertising knowing only the interactive world will find a return to the old ways confusing and perhaps wrong-headed. Others, who have spent some time on both sides of the fence but where never happy with GRPs, audience, and reach/frequency measures of communication impact, also won’t want to see such imprecise, antiquated mechanisms brought to the New World of online advertising. Then there are those who view anything that mediates the expression of media’s effectiveness — in any way other than that of ultimate sales — as simply smoke and mirrors.
Well, there is validity to each of those viewpoints. I mean, let’s face it: Until the Web, media measurement has been more artistic statistic than scientific exercise. But I don’t want to get into that now (there is a small book to be written on that subject). Rather, I want to address the two forms of expression from traditional media that may help lead us to the Promised Land of online becoming part of the media mix.
Audience
Much of site selection — much as in print selection — is finding an audience that is assumptive, based on content. But, unlike print, currently I can’t buy advertising against a rate base, which guarantees a certain potential for reaching a magazine’s audience.
I can determine a target and see if it resides “within” a particular vehicle. I then buy advertising against an actual demographic and receive guarantees for delivery against that demographic.
It may be difficult for most sites to start delivering against demographically specific audiences, save for maybe those that have registered users who volunteer that kind of information. But sites should be able to start guaranteeing rate bases, just as print does. With the kinds of technologies available now for ad serving, they should be able to institute certain controls that can allow for the delivery of an advertiser’s message to a given audience size, and they should be able to sell inventory based on guaranteed rates.
There’s no reason the process cannot still be based on a cost-per-thousand price structure, as it is with print, but there should be more accountability on this level than there is today. Through something as simple as frequency controls and cookies, it wouldn’t take much for sites to offer this kind of inventory structure.
The problem is that there is so much more monetized potential in the waste impression-only purchasing once promised, publishers would be reluctant to let this kind of inventory structure go.
But I would argue that if sites could deliver an actual audience or a rate base, more advertisers would be willing to move dollars into the online medium. And it would help place an online campaign within the context of the rest of an advertiser’s marketing effort.
GRPs
GRPs represent a unit of measure that is 1 percent of an audience — not necessarily a unique 1 percent, but one unit consisting of 1 percent. The way GRPs are calculated in traditional media is as follows: (Impressions/Universe) x 100 = GRPs. The universe in this equation is basically the audience you are reaching. GRPs are usually considered untargeted (targeted GRPs are called TRPs, or targeted rating points).
So, if I buy 100,000 impressions on a site that gets 1 million visitors, that’s 10 GRPs for just that site. But for the Web as a whole? 100,000 impressions are about a .08 GRP (assuming total Web population as reported by @plan). That’s nothing. There isn’t a client in the world who would spend money to buy just .08 of a GRP.
This measure is, again, relatable to audience.
If I buy 100 rating points against adults aged 25 to 54, the broadcast vehicle I buy will guarantee that I get my message in front of that audience. (Radio is a bit different, for reasons that the media buying industry has begrudgingly accepted, but that is a discussion we simply don’t have time to get into right now.) The guarantee is shaky, but statistically validated.
If we started dealing with GRP/TRP currency in the online space, coupled with the proper demonstration of ad effectiveness, maybe traditional advertisers would then understand where the medium fits in with the rest of their efforts and so up their investment.
