• by David L. Smith , June 13, 2008

As someone who came out of traditional media and who believes its members have a good understanding of what clients want and need, I am always astounded by the question of the validity of using GRPs for a Web metric. But then I back off and remind myself that there is a whole generation of advertising people who are from the Web era with no background in traditional.

The answer is yes, of course.

Why? Let me count the ways.

1) They already exist, even if they are not used or obviously available.

2) Advertisers who spend most of their money in traditional media (the majority) are used to this metric, along with reach and frequency as a KPI (Key Performance Indicator) in evaluating media plans.

3) GRPs are simple to compute.

4) They don't have to be the only or even primary metric for the Web, but they can be an important data point.

5) If using them makes it easier to sell your plan through, why not?

There are those who say that the Web should go beyond the GRP. Don't use old metrics when you can use new ones, they say. I agree that new metrics should be used wherever possible. But cross-media comparability is important in providing the evidence to a traditional advertiser that makes it easy for them to approve a plan. So, why not use the metric?

There are others who say that they don't see GRPs reported out anywhere, so what the heck are we talking about? Here's a simple Media Math 101 lesson.

A GRP (or a rating point) represents 1% of the audience measured. Traditionally, as TV households represent virtually 100% (actually 98 point something) of the U.S., TV households are commonly used as the denominator in any equation. This then produces an apples-to-apples comparison on a cross-media basis. But you can use total U.S. households as the denominator, too. The numerator is the number of impressions. For your information, Nielsen reports that there are 114,890,000 total households and 112,800,000 TV households in the U.S.

Using TVHH as the denominator, if you have a campaign that has 5MM impressions, it would then have 4.4 GRPs. It's as simple as that. Many clients use the concept of TRPs, which are target audience rating points. The computation for this is a little more complex, as you have to determine the U.S. universe for the target (again, Nielsen is a good source). For target audience impressions, just multiply the total impressions by the percent composition. Divide that into the TA universe and you have TRPs.

While we do not generally have easily accessible reach and frequency data for the Web (that's a whole other article), we do have a number of R/F estimating tools from NetRatings, comScore and the third-party ad servers. While each of these has its limitations, they are helpful in estimating Web reach and frequency for combination with traditional media efforts to demonstrate the overall reach and frequency for your client. Remember that the formula is reach x frequency = GRPs. In the traditional world, this is generally done on a four-week basis for broadcast and a monthly basis for print.

Using the above concepts, it is fairly easy to build GRP computations into your spreadsheets, as you are reporting out the amount of advertising weight that you are purchasing on each site. As we get more and more into video on the Web, it is only natural that the clients will want to understand the potential impact (in eyeballs) that the Web campaign reaches vs. a TV campaign.

In this world, impressions alone do not tell the full story. Don't limit yourself to this metric, but consider using it as one more way to improve the understanding of your client relative to the amount of advertising you are buying. They will a) get it, and b) quickly see that the cost per GRP for the Web is less than that of TV. Yet another reason to use the Web.