September 2002  
Practical Application of the R/F and GRP concepts for the Internet
By David L. Smith
In an earlier piece in this space, I talked about the ARF approved definition of Internet Reach as "the number of different persons exposed at least once to an Internet advertising message over a specific period of time, usually four weeks, but preferably starting with a one-week period."  
It is important to note, once again, that this definition is meant to align the Internet definition with that of other media. Internet audience measurement companies have historically used the reach term to characterize the total site coverage or cume potential rather than use it in the historical context of a campaign metric, not a medium potential metric.

I have had a number of inquiries from both sites and agencies as to the use of the GRP and R/F concepts. In this article, I am going to offer a basic primer as to how to start using these concepts today.

First, sites and agencies should be evaluating the programs in place from Nielsen NetRatings (through IMS/MediaPlan currently and soon to be on Telmar) and Atlas/DMT with their comScore overlay (up on Telmar "real soon now" but available without the comScore overlay directly from Atlas/DMT now). While these software entries do not yet have the overlay of SCM data and, as such, probably overstate reach (see previous articles), they represent a good first cut at the problem.

At the same time as they are evaluating the available offerings, some work can be done to pave the way for agency and client understanding by Interactive media operations.

The first has to do with GRPs. This was described in the work presented last fall at the iMedia Conference by Tim McHale of Underscore and is posted on the iMediaconnection.com site. The computation is really quite simple. (I am going to use an example for GRPs in the US, but this can be done for any geography). First, set up a spreadsheet with the sites purchased and the impressions by site. Where possible, use an audit by I/PRO, ABC or BPA or other reliable source to determine what percentage of visitors to the site are in the U.S. Enter that in a separate column and discount the impressions to the US based impressions. Then, use syndicated research to arrive at a percentage composition against your target audience. (Ideally this is the target used for other media so that all GRPs are comparable). Enter this percent composition into a separate column. Multiply your percent composition by your US based impressions to come to a target audience impression number by site. The last data point is the universe of your target audience nationally. You will want to divide your target audience impressions into the target audience universe, and then multiply by 100 to get GRPs. In a formula, it looks like the following:


((Site impressions x % US x % composition)/(Universe))*100=Target Audience GRPs

Some agencies call these TA GRPs and some call them TRPs. Same thing, no matter what they are called.

Once you have the calculation, include it in your all media flow charts. The first thing that will probably stand out is the small number of GRPs in Interactive media vs. other media. This realization is what we call a “good start”.

Relative to R/F, there is a way to reflect R/F in your buys today, even if the ad servers are not ready to execute this way.

Let's use two extreme examples: The first is a DR product and the second is a product with a branding goal (we recognize that many Interactive campaigns are a little of each).

For the DR offering, a lot of research has indicated that there are diminishing returns after the first impression. So, instead of buying 1MM impressions, buy 1MM impressions with a frequency cap of 1 (or 1MM uniques). This controls frequency and gives you a reach-oriented execution.

For the branding offering, a lot of research has indicated that the sweet spot is 4-7 impressions. For sake of simplicity, let’s set the frequency goal at 5. For the same 1MM impressions, buy 200,000 uniques at a 5 frequency cap per site. Sites may not be able to deliver 5 impressions for all 200,000 so a slightly larger number of uniques may be necessary to deliver 1MM overall impressions, but we are definitely moving in the right direction with this type of buy.

Of course, we have to lobby the third party ad servers to give us reporting that permits us to track campaigns on the above basis. And we need to learn a lot more about what average frequency happens when we have frequency capping. But the above formula is directionally correct.

To recap:

  • Investigate the solutions out there today.
  • Calculate GRPs for your current campaigns and compare them to other media.
  • Make an attempt to impact the R/F of your current buys through frequency caps and buying a given number of uniques, not just impressions.
  • Help lobby the third party ad serving companies to facilitate this kind of buying.

David L. Smith, President and Media Director of Mediasmith, Inc. in San Francisco is a nationally known expert in the areas of new media application, media strategy and media planning. A thirty-seven year veteran in the advertising media management arena, Smith has a major involvement in national committee work to establish and refine standards in metrics, business practices and financial issues for Interactive advertising with organizations such as the AAAAs, IAB, OPA and the ARF. He currently chairs the Online Reach & Frequency Committee for the ARF.