"Sixteen million," she repeated.
"Really," I deadpanned.
"Says right here in the New York Times. Sixteen million, seven hundred thousand people sat in front of their televisions and watched those celebrities on CBS twirl," she asserted.
"It's an estimate," I said.
"No, it's a count," she replied with conviction.
At this point I had to make a decision. I could not have cared less what she thought. But while waiting on the tarmac before takeoff, she had told me in great detail how she was recently promoted to be the head of all advertising for her organization. Her business card stated she was an EVP with a well known, Fortune 50 company. We had talked briefly about online and on-demand television strategy. Could she really be that ill-informed? I must have misunderstood her. Or perhaps she was joking.
advertisement
advertisement
"Did anyone call you?" I asked with a smile. "I don't take phone calls after eight o'clock," she said.
"So you weren't counted?" I teased.
"Oh, they know that you're watching," she assured me.
"Who's they?"
"Nielsen." She was starting to scare me.
"I wouldn't be so sure about that," I said.
"They get the data from my television, they get data from the others." Now at this point, I had a sinking feeling she might actually be a member of a Nielsen panel -- which would explain a great deal and make me feel a little sheepish to boot.
"Oh, so you are a Nielsen family?" I offered.
"No," she said. "My name is not Nielsen. I gave you my card. It is Swanson," she spat indignantly.
"I am sorry," I began. "What I meant to ask was, 'Did the Nielsen Company recruit you to be a member of the television panel?'"
"I would never participate in such a thing. Don't be ridiculous."
"But I thought you said they were connected to your television--" It was my turn to be confused.
"Yes. I have cable. Nielsen knows everything I watch. They get my television data from Time Warner. Yours comes from Comcast or DirecTV or whoever."
"Really?" I said.
"Yes. This is what I've been telling you." She shook the copy of the New York Times in her hand. "Sixteen million, seven hundred thousand viewers watched 'Dancing With the Stars' on CBS." I decided she was a lost cause. But I could correct her on another point.
"Ahhh," I began. "But you know, the show airs on ABC."
"You want some cream?" she asked, obviously confused.
"Yes. Three creams, please." I quipped.
"What are you going to do with the cream?" she said looking at me as if she had just discovered that I was some sort of criminal.
"I drink it straight," I said. "Calms my nerves."
"That's not healthy," she said. "You're going to have to ask the attendant for it yourself."
"Thank you anyway. I'm feeling better already," I said as I pulled my Bose headphones from my bag and dove back into my book.
While not an official polling, this particular conversation is not atypical of those I routinely engage in. My experiences indicate that most reasonably educated people -- even industry people -- believe Nielsen counts viewers. How is this possible? If you look up the Timesarticle to which my seatmate was referring, you will find the word estimates once, as in "according to Nielsen estimates." At the same time, you will find the number of viewers for a program listed fourteen times, as in "The CW's 'Gossip Girl' earned 3.3 million viewers."
Honestly, what was my seatmate on the plane supposed to believe? There are some who might argue that she should have known better, but the truth is the industry propagates the counting myth. We let sellers get away with posting to terribly poor data and turn up our noses at sophisticated (albeit not fully baked) data solutions from Google, TNS and Rentrak.
At the end of the day, counting is usually better than polling. When data from broadband video is available, even when it is broken into five or ten minute streams, those "tuners" are counted. When Live+ numbers are crunched, those "viewers" are estimated, and I would argue that the estimate is more like a SWAG. If the television industry is to thrive, measurement is essential. But poor measurement is worse than no measurement at all. Just ask the radio guys.