As prime-time television enters its ``second season,'' when several programs give way to midyear replacements, a little black box monitoring the TV habits of 2,400 American households has set the national networks' teeth on edge. These new ``People Meters,'' developed by the A.C. Nielsen Company and put in place at the beginning of this year's TV season, are telling ABC, CBS, and NBC something they vigorously dispute: that 10 percent fewer people are watching their prime-time shows.
``We weren't really expecting such a drop, more like 2 or 3 percent,'' says Gerald Jaffe, research vice-president at NBC, the network least damaged by People Meter ratings so far. It lost 4 percent of its audience during prime-time hours, and an average of 20 percent during the Saturday morning cartoon period.
But CBS and ABC both lost 13 percent during prime time, about six or seven points more than they had expected. The bottom seems to have fallen out of Saturday morning cartoons as well, a spot that pulls in about $150 million in advertising for the three networks. Nielsen's new system clocks at least 20 percent, and possibly 50 percent, fewer children tuning in.
For some media-watchers and advertisers, this sudden ratings drop underscores their contention that the three big networks have been losing a share point each year since 1978, says Stephen Fajen, a senior vice-president at Vitt Media International Inc., a media-buying firm in New York.
Some think this will continue as cable stations and independent channels multiply and the expanding home video market keeps growing.
But there is also a certain sense that ``whatever the chinks in their armor, the three big networks are still the most effective way to reach a mass audience,'' says Betsy Frank, senior vice-president of media research at Saatchi & Saatchi DFS Compton, a large New York advertising agency.
What bothers the networks most, of course, is that the People Meter ratings message has taken a bite out of their advertising dollars. Since CBS and ABC underestimated their own cut in audience share (NBC was a bit closer in its estimate), they are having to pay a number of ``make-goods.''
Make-goods are free units of advertising time. Because the networks guarantee ad agencies that buy ``up front'' time (spots bought for the entire year usually come with a network guarantee) higher ratings than the ones now coming through, the networks now have to make up for lower ratings by offering free units.
``They're losing out, since the giveaway units tend to be scatter time [when advertisers buy at the last minute, and only for a short period of time], which usually make more money than time bought up front,'' explains Ms. Frank at Saatchi & Saatchi.
If the ratings stay this low, or continue to slide, the networks may have to lower their beginning rates, she says. ``If the meters are showing 10 percent lower ratings, the networks have 10 percent less inventory to sell.''
But Mr. Fajen, at Vitt International, which buys mostly local, cable, and syndicated time, thinks the networks will continue to overcharge advertisers ``even if ratings settle around 9 percent,'' as long as the demand is high.
The networks insist demand is higher than the People Meters say it is.
``Nielsen's local sweep didn't show the same declines,'' asserts Jerome Dominus, vice-president in charge of sales at ABC.
The meter has a button for each member of the carefully selected household and is preprogrammed with their sex and age. Every time a person turns the set on or off, or comes into or goes out of the room, he is supposed to press his button.
NBC's Gerald Jaffe argues that the amount of effort required by this method automatically alters the nature of the sample, because he contends it concentrates on only that small segment of the population that is willing to participate in such a survey.
The People Meters are also a big nuisance, according to many network executives.
``The person who does [the old survey] only has to do it 13 weeks a year,'' Mr. Jaffe says. ``The People Meter person has to do it 365 days.'' And ``when you have guests,'' he says, ``they have to be logged in, too.'' Because of this, he and his network associates say they can't be sure every member of the household actually pushes the buttons. Many, like Ray Warren, Eastern sales vice-president at ABC, do not believe ``a three-year-old child will remember to punch out when they leave the room to play.''
The networks also charge that the sample survey is skewed toward cable viewers. They claim only 26 percent of the population have pay cable, while 33 percent of the People Meter households have it.
``There are extra options in the sample that more Americans don't have,'' Mr. Dominus says.
Nielsen defends the precision of its boxes. Follow-up phone interviews since September agree with the meters' numbers, the company says.
Though Nielsen, which has so far managed to hang onto its ratings monopoly, won't say there aren't potential problems with the People Meters, it insists that previous testing shows the little black box to be more accurate than the old system of diaries. The diaries system automatically registered when the television was on, but a member of the household was expected to keep a diary of who watched what at the end of each day, or week. After two years of parallel testing, People Meters were adopted as the official system in September because of complaints that the 37-year-old diary system was outdated and imprecise, Nielsen spokeswoman Terry Luke explains.
``If you're the diary keeper, and you have a couple of kids who watch a lot of TV, at the end of the week you're going to have to write something down. How much of what went on are you going to remember?'' Fajen asks.
About 50 percent of American homes have cable, and between 28 and 30 percent have pay cable, says John Dimling, a Nielsen senior vice-president. This means that Nielsen's new system is not overrepresenting cable-watchers. ``A little over 50 percent of our sample are cable subscribers, and about 31 percent have pay cable,'' Mr. Dimling says.
The fact that the meter is not showing a decline in overall viewing also indicates ``people are pushing the buttons.''
The networks stress, however, that they're not contesting the accuracy of the numbers recorded. They just think the numbers are not complete.
``There must be an awful lot of people who aren't handling the equipment properly,'' Mr. Warren says.
While some network executives see some advantages in the People Meters, many believe that Nielsen adopted the system more in response to threat of competition by a British research firm called Audits, which also offers the new rating system. Some advertisers had talked about using the Audits figures to defend their contention that the networks were overcharging for commercial time.