I was complaining to someone recently about a particular new show I didn’t like but how it makes sense that the network picked it up, because it totally fit with the brand and was paired with other shows that I also really don’t like but are very popular, and the ratings bore out that this was a good decision because it did well. And he asked me how ratings work. Like, how do they know what he’s watching every night? And I forgot that this was not common knowledge.
So: first up, they don’t know what you watch every night, unless you specifically tell them. Some companies like to keep track of what’s trending, so if you use a hashtag on Twitter, for instance, they might keep track of that. But that’s not what we mean when we say “ratings.” (It is what we mean when we say “buzz,” which is an actual thing. If a show has good buzz, a lot of people are talking about it on social media. Real thing.) Ratings are almost always generated by the Nielson Company. The Nielson Company takes the number of households in the country and breaks them down into demographic groups based on age, gender, income, education, geography, etc. Then they reach out to random people looking for households that fit into the various demographic groups. They get a good sample size for each group, and the households in their samples get a box they put on their TV that records information about what they’re watching and sends it back to the Nielson Company. This is not done in secret. The families that are chosen are given a little money in compensation. (It used to be that families would have to fill out diaries of what they watched on TV, but now digital 2-way communication is common.) The results are then projected out to the rest of the population: In other words, if 2% of the households surveyed by Nielson watched a particular program, the assumption is that 2% of households in the country watched it. (This is also, I believe, how ratings work in radio.) The ratings will be broken down by demographic, which is how we know that The Good Wife did well with high-income households.
I recently got home to a letter from the Nielson Company with a dollar bill, which the letter explained was a down payment on my agreeing to answer a phone survey. If I answered my home phone, which I never do, it would have been a very short phone call because the first question is “who is your employer” and my answer disqualifies me from the rest of their survey. (I still used the dollar to buy a candy bar, though. Free food is free food.)
In recent years, DVRs have become more and more common and it’s increasingly uncommon for people to watch shows live. Nielson has compensated by putting out several sets of ratings: Live, Live+1, Live+3 and Live+7. These measure how many people watched the show live, within 24 hours, within 3 days, and within a week, respectively. Advertisers hate anything that isn’t live because people often fast-forward through commercials, but networks are hard at work convincing them those viewers count just as much, because they have to be staring intently at the screen to know when the commercials end and they tend to be the ones who watch so much TV that they can recognize a given commercial and the message with just a couple frames. Last spring, the CW announced a deal with Netflix where current seasons of their shows will be available for streaming 8 days after the last episode of that season airs. The AV Club thought 8 days seemed rather arbitrary, but it was because the CW wants to get the Live+7 numbers. The CW also puts a lot of care into measuring how many people stream their shows, which they sell as their demographic being young, the long-standing assumption being that young people are more “brand-flexible” (willing to try new things) and have more disposable income than people who actually think of themselves as adults. Thus, young people are more valuable to advertisers selling cell phones and personal care products. (This is as opposed to CBS, which typically sells itself on household numbers instead of key demo numbers because they do very well with older audiences. To their credit, older audiences are more valuable to advertisers who are selling prescription drugs and pickup trucks.)