My dog has a rubber toy that has a hollow core, and what you’re supposed to do is fill the core with peanut butter or cheese and watch as she nudges it and bounces it and tries to get at the treat inside. Mostly, though, she just stares at it with a totally baffled expression.
It’s the same expression that people in the television business have when they’re staring at the Nielsen ratings. Somewhere buried in the cross-tabs and demographic breakdowns is an argument for declaring their television project a hit. They just have to figure out how to spin the numbers.
“These are very good numbers,” someone from the studio research department told me a few years ago on the morning after a show I was working on had its second-season premiere. “You’re up in the 25-to-54 range, you hold your lead in, and year-over-year you’re showing growth for the network.”
I had been staring at the ratings breakdown for nearly an hour when he called, and I couldn’t quite figure out how the show had performed. He saved me a lot of agita.
Or would have, had someone else from the network research department not called moments later to say: “You had a soft debut. You did okay. The low-middle range of okay. We’re concerned.”
That morning’s whiplash was produced by the audience-measurement data of the Nielsen Media Research company. Through a variety of complicated audience-measurement tools—and, I suggest, a certain amount of witch-doctoring—the Nielsen ratings come out each day to tell the advertisers, networks, shareholders, and executives what America watched the previous evening.
There are robot telephone polls, live calls, viewing diaries, set-top boxes, and of course the famous “Nielsen Families”—households that allow Nielsen to measure directly what they’re watching—all collecting information about specific audiences, which is then extrapolated into a snapshot entitled “What Happened Last Night.” If it sounds a lot like political polling, that’s because it is, including the gripes about methodology and “skewing” and “sample size” that the previous night’s losers always engage in. If you could sit in the writers’ room of a show that had a disastrous night in the ratings and close your eyes, you wouldn’t know if you were among coastal elite electric-car-driving Hollywood liberals or MAGA-hat-wearing red-staters. Everyone, it turns out, hates the pollsters.
“Have you even met a Nielsen Family?” We ask one another as we digest the ratings. And of course no one has.
One day, though, I did.
I had just finished a book tour event at a bookstore in a university town. As I was gathering up my stuff and wondering where the closest drink could be found, a young man approached me and told me that he and his roommates had been asked by Nielsen to have a “ratings box” attached to their television. He and his roommates—all males between 22 and 29—were about to become a Nielsen Family. They were about to help determine the national television ratings for a key demographic group. Their specific viewing habits were about to be extrapolated and treated as representative of their cohort as a whole, which was interesting to me, because I had a new comedy about to premiere.
“I will pay you $15,000 apiece,” I said, “to watch my show.”
In my mind. I said those words in my mind. Because somewhere in my dim memory I vaguely recalled signing a paper promising not to do that very thing, swearing some kind of contractual oath not to game the system, when I signed my studio contract. And though I would have dearly loved getting a celebratory and unequivocal call the next day from the studio and network research departments, I kept the exchange on the ethical high ground and no money was exchanged. I simply asked him to pretty-please watch my show and, if at all possible, to leave the television off until just before my show came on and to turn it off immediately after the studio logo appears in the end credits. What I’m going for, I told him, is a noticeable bump.
I don’t know if it worked, though I have to admit that the series had a pretty impressive debut, especially in the coveted male 18–35 demographic. But this kind of thing just adds fuel to the rising chorus of voices in the television business saying that the Nielsen monopoly on audience measurement needs to end.
The Video Advertising Group is a trade organization made up of cable-television producers and old-style broadcasters that has accused Nielsen of undercounting the broadcast and cable audiences by nearly 6 percent. They have forced the company to undergo an audit of its methods and measurements, but the position of the companies is clear: They insist their audience share is being undercounted. They are convinced more people are watching their shows than Nielsen is measuring and that rather than being in the middle of an alarming audience free fall, cable and broadcasting audiences are robust and growing. They are looking at the Nielsen data like my dog, absolutely certain there is a cube of cheese in there somewhere.
The streaming services have it a lot easier. Nielsen attempts to measure their audiences, too, though it’s more for bragging rights than anything else. Netflix and Hulu and their competitors have only one key metric to report—new and renewing subscribers—and those are numbers they announce quarterly, on earnings calls with investors. Those are the only “ratings” that count.
But their utter disdain for the grubby cable-and-broadcasting audience measurement must be galling to those ad-supported businesses. Netflix, for instance, regularly releases what it calls “datecdotes”—a little bit data, a little bit anecdote, get it?—filled with pointless factoids about the “Most Popular Show on Thursday Among Teens” and the “Most Talked-About Limited Series” on the service. None of these “datecdotes” have any real measurements attached to them. It’s just the rich, cool streaming kids making the broadcast and cable nerds feel clumsy and broke.
What the Video Advertising Group wants, it says, is more “accuracy.” If that were true—and it isn’t—they would be the only part of the entertainment industry universe that actually wants to know, down to the person, how many people are watching its products. Accuracy in this arena serves no one’s interests. They don’t want it right. They want it 6 percent higher.
The studio that produces the television show wants a large audience so it can charge the network a little more the following year. The network wants a large audience so it can charge the advertisers more for each 30-second spot. The advertising agencies want to report that their brilliant creative messages have been seen by an enormous audience, which will allow them to keep their clients and attract new ones. The consumer-products companies that buy time on television want to report up the chain that consumer awareness is high, the marketing team is on top of it, and that everyone deserves a raise.
So at the end of the current battle between Nielsen and the television business, the result will be a kind of affirmative action for Old Television. After some adjustments here and there to its methods and some statistical tweaks, the recalculated audience for cable and broadcast television will be found to be 5 percent higher than previously measured.
Not 6 percent, because you have to leave something on the table for later. You always want to have a treat in your pocket, as all dog owners know.
We want to hear your thoughts about this article. Click here to send a letter to the editor.