by Dave Zornow
Cable Avails, July 1998

USA Networks led all basic cable networks with a 2.6 prime time rating in the first quarter. At the end of March, they ran their Nielsen Galaxy Explorer system to document their success before distributing the story to the press. Owning the highest prime rating in cable would seem to be a non-controversial press story which would stand up on its own merits. That is, until Nielsen's printed reports came out and contradicted the USA press release, reducing the 2.6 rating to a 2.5.

Which number was correct? 

According to Nielsen, they both were. Tim Brooks, executive vice president of research at USA Networks, raised the discrepancy to Nielsen's management saying that the dual truths undermined the value of USA's victory and the credibility of all Nielsen numbers. Brooks was further perturbed to find that in both Nielsen sources, the audience projection in thousands WAS THE SAME although the rating was different. Nielsen's Galaxy Explorer produced a 2.6 rating and 1.854 million average prime time households, but the Nielsen Cable Activity Report (NCAR) and the Network Cable Audience Composition book (NCAC) showed a 2.5 rating and 1.854 million.

"Tim is correct in saying there are some differences," says Sara Erichson, vice president national marketing of Nielsen's cable division. "Explorer is our best and newest system. There are minor differences in how the numbers are rounded between Explorer and some of our older products," she says. "Some of our older systems are in the process of being rewritten or phased out - it is in these systems where you will see differences of a tenth of a rating point when compared to Explorer."

USA Network was not very understanding of the Nielsen nuances. "The press wants to know what the ratings are when the quarter ends, and they are not going to wait three months for Nielsen to polish and buff their numbers," says Brooks. "The first number produced had better be the right number. Unfortunately, that number, when produced from Nielsen's Galaxy system, doesn't match what the [subsequent] printed reports show." Brooks notes that sometimes differences can occur due to different definitions and specifications used to generate the results - for example, including or omitting Nielsen's "non-report row weeks" or different start and end times for prime time. But the first quarter fiasco didn't involve any such subtleties. Two numbers that should have been identical simply weren't.

Brooks thinks that Nielsen's response is a symptom of a greater problem. "Senior management can easily the difference between a 2.5 and a 2.6," he says. "When you say both are right and tell them Nielsen stands behind both numbers, they don't accept it." Brooks thinks this is an example of how Nielsen fails to see the cable business from their customers' perspectives. "Their behavior is in stark contrast to how we have to run our business," he says. "We need to make decisions based on firm recommendations. Nielsen's clients can't get by the same way they do, saying both of these numbers are right."

More than a USA Problem

Prevue Network's 1st quarter 1998 awareness program earned it a .5 rating in their pocketpiece, with increased exposure among Teens and 18-34s. Or did it? According to Nielsen's Audience Composition Report, Prevue pulled a .4 - which is less than the previous quarter. Does rounding make a difference to Prevue? Which would you rather report to management - sustained ratings, or a 20 percent ratings slide?

FX also showed discrepancies in their TV HH projections between various Nielsen sources. In their pocketpiece, FX's 1.0 rating in prime time is worth 336,000 households. But in the NCAC, a 1.0 is worth 1,000 more households. USA's Monday-Sunday 7a-1a estimate shows similar variations with their 1.3 rating projecting to either 926,000 or 927,000 TV HHs depending on which Nielsen report you consult.

Travel and Goodlife TV, two smaller Nielsen-measured networks, disappear entirely depending on the source. The Travel Channel's prime .2 rating is worth 36,000 TVHHs in the Audience Composition book, but is just "chicken feet" (Nielsen's nickname for the << symbol which means it's too small to report) in the NCAR. GTV delivers a .6 and 29,000 in the NCAC book, but for some reason it's not reportable in the NCAR.

Is all of this really important to how the cable network business is conducted on an average day? "I think there are bigger fish to fry," says Bob Sieber, vice president audience development at Turner networks. "It's a long standing issue which goes back to when many of these NHI programs and reports were originally created," he says. "The problems a few years ago were horrendous - it's not as serious an issue as it once was." Sieber thinks that the problem could be remedied if Nielsen created guidelines for their clients and their systems, especially for press use. "They need to develop a style guide for use of their numbers, similar to what newspapers and magazines use to assure consistency," he suggests.

Nielsen's Erichson is sympathetic to the network's complaints and is developing a plan to address these inconsistencies. "Our first step is to conduct a comprehensive review of all of the sources of difference, and then identify and explain them to our customers," she says. Although the first step should be completed over the next few months, the follow up may take longer. "If we are about to retire a system, we have to ask ourselves, 'Does it really make sense to change it?' A better place to start is to understand where there are differences and why, and then develop a timetable to address the problem." ##

Dave Zornow is President/TNG Research, a media research consultancy and applications development company that works with media sellers and research providers.

This article was originally published in July 1998 by Cable Avails magazine.

Return to: Home • Published Articles