Setting Reasonable Goals a Key to Planning in 2004

Jan 19, 2004  •  Post A Comment

In January it is customary to list resolutions for the new year. For media planners, a number of common practices exist that are problematic from the standpoint of the underlying research. In this season of new beginnings, it is important to recognize them and resolve to avoid repeating them. Thus, my resolutions:
Resolution 1
I will evaluate the reach and frequency of media plans on a four-week, or at most a quarterly basis-never annually. All of the underlying research is based on these shorter periods. Nielsen’s Persons Cume Study, which is the basis for the most commonly used TV reach/frequency systems, covers four weeks in November/December 2001. People Meter sample turnover makes unification beyond three months problematic.
Magazine reach/frequency is derived from the number of issues read out of the past four, yielding at most a four-month base. Out-of-home is typically measured in 30-day increments. As a result, from a research standpoint, 52-week reach estimates are spurious. They are also irrelevant for most consumer behavior.
Resolution 2
I will remember that advertising exposure in each medium is measured differently. Adding 100 gross rating points of television to 100 GRPs of magazines yields the mathematically correct 200 total points, but that number is meaningless from a communication standpoint. The process is not unlike mixing a pound of salt with a pound of filet mignon. The result is 2 pounds of food, but nothing you would want to eat.
Resolution 3
I will consider Nielsen the primary source of television audiences, and will look to MRI, Scarborough, Media Audits and other services only when I need supplementary data on the viewing habits of product users, retail shoppers and other market segments not measured by Nielsen. Since all buys will eventually be executed in terms of the Nielsen ratings, the support for television recommendations should come from that source.
Resolution 4
For the same reason, when working with television plans I will use Nielsen Universe Estimates for national and designated market area demographic populations by age, sex and household characteristics, including Hispanics and blacks.
Resolution 5
I will express the reach of a media plan in whole numbers as a way of communicating to my client that the figure is just an estimate. Mathematical models that calculate reach have a margin of error of several percentage points. Computer systems that carry the calculation out to tenths or hundredths of a point (e.g., 73.68 percent) give a false sense of precision.
Resolution 6
I will remember that radio reach builds differently from television. With TV, a given number of points in a daypart will yield a predictable reach. With radio, reach is determined by both the GRP level and the number of stations that are bought. Within limits, a planner can specify how much reach is needed and the buyer will respond by scattering the spots over many stations to build more reach, or concentrating them on a few stations when less reach is called for. For general demographics, such as women 25 to 54, radio reach in excess of 60 percent to 70 percent is hard to achieve because of the need to buy many small, inefficient stations.
Resolution 7
I will reserve the words “optimal” and “optimize” for those times when a recommendation is based on a computerized optimization run. Unsupported generic statements such as “This plan will deliver optimal levels of reach and frequency” sound good, but in fact are meaningless.
Resolution 8
I will remember to consider the geographic area covered by the estimated reach and frequency of a media plan. To be completely accurate, plans that have national and local heavy-up television weight should be reported three ways: (1) delivery against people in the nonspot markets who will see only the national venues (network, cable and syndication), (2) delivery in spot markets that reflects the combined weight of a local buy and national spill-in, and (3) the consolidated national delivery that reflects the population-weighted reach of target viewers in spot and nonspot markets.
Resolution 9
Finally, I resolve to remember that media planning is an inherently judgmental process. It should not be guided by simplistic mathematical models such as the “point of diminishing returns” that reflect marketing and communication assumptions that cannot be supported.
Like most new year’s resolutions, I expect many of these will be honored more in the breach than in the practice. But hopefully, planners will keep them in mind as they go about their day-to-day business.
Roger Baron is senior VP, media research director, for Foote, Cone & Belding.