What We Know About Frequency in TV and Digital Advertising: Pt. 1

In this brief series the three authors combine their respective expertise in TV and digital media research to share what they have learned about campaign reach, frequency and sales impact to provide the media industry at large ways to think about these marketing metrics as we transition to a cross-platform reality.

An Examination of Frequency in TV and Digital

While TV has always been known as "the reach medium" for its ability to deliver audiences in the tens of millions in a single half-hour program, one could also argue that it has historically been the frequency medium as well. TV still reigns supreme as the king of media engagement with the average American viewing for 5 hours and 9 minutes per day according to Nielsen (not including digital streaming/OTT or VOD), which is about 50% higher than the total digital media consumption reported by comScore. As a result of so many people spending so much time viewing TV, advertisers have the ability to design media plans that they can be assured will reach a high enough percentage of their target audience in addition to optimizing the number of times the average viewer is reached with an ad. However, due to variation in usage level across the population, even an optimized schedule in today's nonaddressable TV world will have large numbers of viewers reached too many times and large numbers reached an insufficient number of times.

Today's increasingly cross-platform media paradigm demands a more unified view of TV and digital, and this begins with a deeper understanding of the essential building blocks of marketing and media planning: reach, frequency and sales impact.

How Frequency Operates on TV

It used to be very simple. For about 210 GRP per week you could get a 70% reach and an average frequency of 3. Many (but not all) pieces of evidence pointed at 3 per week as being just right for the average mature CPG brand. Herb Krugman had logically derived the idea of a 3 frequency and then backed it up with brainwave research on his secretary. Mike Naples did a cumulation study that pointed to a 3 frequency, then Bill Harvey did one ten years later and got the same thing. TiVo Research/TRA findings for CPG based on actual sales effects have shown 3-5 to be the typical optimal frequency per week.

Today the same number of GRP in a week will tend to get you ~45% not 70% reach. This is the effect of lower ratings on reach. Higher ratings cause increased verticality in the reach curve. Because the decay in terms of sales effect is so steep, you really need to maximize reach of the ROI Driving Purchaser Target in each two-day period -- because you never know when they will go shopping and you want a frequency of two in the last 48 hours before they go shopping (CPG).

The buying culture does not really permit this degree of rationality to be applied, say, in the Upfront. The theory used to be that you could buy sloppy but cheap in the Upfront and then fix and optimize in the allocation of brands to the programs you had bought. Results of course are always better when one applies discipline at every stage, not just select stages. Today the popularity of the idea of "programmatic" is a long pent up reaction to the irrational aspects of ad spend to date.

The big problem with TV, as many have pointed out in the past, is that the variation in TV usage by household and person is so extreme that the non-addressable medium puts too many impressions where you don't want them, piling up on heaviest viewers whether or not they are all such good prospects.

The opinions and points of view expressed in this commentary are exclusively the views of the author and do not necessarily represent the views of MediaVillage.com management or associated bloggers.