Sponsor (Jan-June 1953)

leeded: a way to project ratings SUA may propose improved Hielhotl. Several agencies ail ready trying it Suggestion: Before reading this article, take a look at descriptions of the old and new in projection methods on page at left. They'll give you a quick refresher course on the problem at hand by Miles David gg om can \ mi projecl .1 rating thai non-projectable? That's the question which this month has sent timebuyers, media directors, researchmen in agencies all over the country into incessant slide-rule-andscratch-pad sessions. With Nielsen Coverage Service and Standard Audit and Measurement Service data finally in their hands, agency people now have to make up their minds how to use it. At agency after agency contacted by sponsor the decision had not yet been reached on how SAM or NCS would be used in projection of radio ratings to get a cost-per-1,000. This is an allimportant policy decision to make because on it may depend ( 1 1 which stations get spot radio business in the campaigns of the next few years, and I 2 ) how much business spot radio as a whole gets in the tug of war with other media. Hanging in the balance is the way in which an estimated $135 million-plus in '53 national spot radio billings will be channeled into the business during this year. You can start a statistic-slinging freefor-all at almost any agency when you get into the method to be used in projecting ratings. But curiously almost every media specialist and timebuyer is agreed on this point: Projecting ratings is ivrong to begin with. Everyone knows this and everyone does it anyway. It's done because it's hard to get a measure of one station's performance in comparison with another unless you determine its cost-per-1,000 homes for the entire listening area. Getting the cost-per-1,000 for just the home county of a station isn't enough for then you often have no way of determining differences between stations. Most stationin a market are going to show you pretty much the same intensity of coverage right within the shadow of their antennas. And yet all stations don't have the same rate. You have to go outside the home area to find out if an announcement costing $100 is worth 10 times one costing $10. But projecting ratings to the outsideof-town counties is wrong since the ratings are usually made on the basis entirely of phone calls or interviews with people inside the town. They are representative of listening only in the area they sample and are published with that specific instruction. What's the solution to the problem? Answer: none. Since sampling people regularly beyond the immediate area of a station is too costly for most broadcasters' pocketbooks, you're not going to have projectable local station radio ratings for the large majority of st?*ions in the near future — if ever. So the practical buyer has to come up with some makeshift answers. Even if buyers didn't want to use cost-per-1.000 as a yardstick, dollar-minded advertising managers would force them to it. It's to the credit of most buyers, incidentally, that they resist over-emphasizing these makeshift cost-per-l,000's in deciding what to buy. They use them as but one of many factors that go into an intelligent choice. (Many feel the only yardstick should be cash register results such as are demonstrated by ARBI. instead of ratings and cost-per-l,000s.) One of the few agencies which has lli us far tentatively decided on the projection method it will use with coverage data to get a cost-per-1,000 is Marschalk and Pratt. The method they're trying I and it may be in use at other shops not surveyed by sponsor I is due to come in for considerable attention in the next few weeks. A Station Representatives Association committee was on the verge of deciding to advocate this method, which they came upon independently, as the best available at sponsor's presstime. Rep salesmen and station promotion, therefore, may be talking it up soon with considerable fervor. The method is a refinement of the use of SAM or NCS total audience figures as a base for projecting a rating. ( Definition : Total audience figures represent all homes tuning station on a daily or a weekly basis, with day and night given separately. These figures are printed on the coverage booklet. 1 The trouble with simply using the NCS or SAM total audience figure as a base for projection is that this shortchanges radio stations and radio as a medium. So what, you say — as long as all the stations are getting shortchanged on an equal basis. The answer to that one is that inevitably cost-per-L000's are going to [Please turn to page 111 ) How do YOU project the unprojectable? sponsor invites timebuyers, media specialists, research men, station operators to submit their own suggestions for making projections of ratings to the coverage areas of stations. Your thoughts on the subject will be printed with or without your name and affiliation, as you prefer. The problem is complex and everyone in the business will benefit from the fullest airing of ideas. So please make the pages of SPONSOR your forum. 18 MAY 1953 37