Occasionally you’ll see or hear someone derisively referring to a TV news story they don’t like with a phrase like, “must be sweeps week.”
Of course, anyone who’s worked in TV knows there’s no such thing as a “sweeps week.” There are ratings periods that are broken up into four-week periods, but even that data gathering is now year-round in every market. Viewership is under continuous measurement.
Previously, ratings surveys in most markets were conducted only four times per year (November, February, May, and July). Though ratings are now issued for every month, some still continue to focus on traditional ratings months, and that is what people are referring to when they talk about a “sweeps month.”
There are two major TV ratings companies, Nielsen Media Research and Comscore. Nielsen is the longtime provider, while Comscore has offered an alternative in recent decades.
Some stations subscribe to both services, some choose one or the other, and a few use neither. It’s important to understand that stations which don’t subscribe to ratings are still affected by them because ad agencies and major buyers subscribe to the data and use it to make their decisions.
Ratings have been traditionally the most often-cited metric. A rating is the percentage of all households that are watching a particular program. That means households that aren’t even watching TV still count against the total. So, if there are 100,000 households in the market and 1,000 of them are watching a program, it has a 1 rating.
Share is the percentage of households who are watching TV right now, who are watching a particular program. If there are 100,000 households but only 50,000 of them are watching TV, and 1,000 of them are watching a program, it has a 2 share.
Impressions have been cited more often recently, representing an estimated number of actual people to more easily compare with digital (online) metrics.
Demographics are the breakdown of viewers by age and gender, and presented as ratings, shares, and impressions. There are thousands of ways to slice and dice the numbers through demographics. By contrast, the word “households” is used to refer to the general overall rating for all people.
You can come up with even more numbers by comparing rating or share growth/loss from book to book. That mean it’s possible to make claims about major audience growth even if the overall audience is still lower than competitors.
You can further break down the numbers by picking which days of the week you are counting. Some local newscasts are often cited as an overall Monday-Friday average but stations may also break down numbers to Sunday-Thursday, Sunday-Friday, or Monday-Sunday, depending on what number is most impressive.
So, how are ratings measured? As a veteran of news middle management, I know a lot about it, but the information is proprietary and cannot be publicly released.
Nielsen’s website explains, “electronic measuring devices and millions of cable/satellite boxes are used to provide local market-level viewing behaviors.”
Comscore’s website says that it measures 75 million TV screens in more than 35 million households, including set-top boxes “from every major cable, satellite, and vMVPD provider.”
Bottom line: There’s a lot of data which can be interpreted in many different ways. Read the numbers carefully and be cautious if someone touts a change in a narrow demographic between two seemingly unrelated months.
Jon Ellis has worked in small-market TV news for 20 years in assistant news director and producer positions.
LINK: More Newsroom Notes