Nielsen ratings are well-known as the audience measurement system for the television industry. However, the actual details of how ratings are calculated are less understood.
Questions like these don’t get answered very often:
- Who are the “Nielsen families,” and how are they chosen?
- How are their choices translated into major industry decisions–such as which TV shows get to stay on the air?
- And, perhaps most topically, are Nielsen ratings still relevant when the way we watch TV has changed so much in the 21st Century?
If you’re curious about the answers to these questions about how Nielsen Ratings work, let’s dig in.
Who Gets to be Part of the Nielsen Family?
Nielsen households are chosen at random, but “random” inaccurately describes the carefully designed statistical procedure Nielsen uses to select these households. The “sample” households are chosen to represent important parts of the total population, drawn from U.S. Census data as well as existing address and telephone database records.
The 25,000 or so homes that contribute data to Nielsen represent about 50,000 other similar households. “Similar” refers to population and income distribution that is broken down into distinct geographical markets and demographic data (age, gender, and race). These elements are of crucial concern to programmers and advertisers, which want to target their products to a specific audience. For example, a young, single, affluent Hispanic female in Los Angeles is likely to have very different viewing and buying interests than a pair of retired, middle-class white grandparents in rural Wisconsin.
Every August, Nielsen recalculates the number of households and adjusts its numbers accordingly–taking some households out of the sample, and installing new households in their places.
Becoming a Nielsen Family
Nielsen sends a letter to a prospective household, explaining its methods and the importance of being part of the sample. If the household is part of the diary sample (see below to learn more), this may be the extent of the contact; otherwise, a Nielsen recruiter will also call or visit the home (an important step in assessing the house for the installation of equipment, if necessary). Nielsen offers a small “thank you” gift to each household in return for participation–not enough to be considered “paying for ratings,” but some recompense for the family’s time and trouble.
There are a number of ways that a household may prove unsuitable for the sample–discovering that a family member works in the TV industry, for example. The household may also decline to participate, which forces Nielsen to contact an “alternate” home on the list. Because of the statistical exactness of the process, the first home chosen is the most desirable, and “going down the list” may cause the sample to deviate slightly from ideal demographic representation.
Metrics and Definitions
Nielsen metrics help determine the size of an audience for a particular TV show; TV networks and stations use these metrics to set their advertising rates. Thus, the rating of a TV show can have a direct effect on the financial bottom line for a network or station.
The most commonly mentioned Nielsen metrics are rating and share.
A show’s “rating” is a percentage of households with TV sets that are watching a TV show at any point in time. For example, assume there are 100 million U.S. households with TVs. If 30 million watched the Super Bowl, then the “rating” would be 30 (30,000,000/100,000,000).
“Share” measures the percentage of households watching a particular TV show out of every household with a TV turned on. Using our Super Bowl example, if 70 million households have their TV on, then the “share” for the Super Bowl would be 43 (30,000,000/70,000,000).
Below are other important metrics reported by Nielsen:
- Cume (short for “cumulative audience,” also known as “reach”) measures the number of different people or households who have been exposed to a show or commercial at least once in a given period of time. Cumes help advertisers determine how many people hear their marketing messages.
- Frequency measures the average number of times a person watched a given TV show or advertisement during a specific time period.
- Gross Ratings Points (GRP) show how big the viewing audience is–no matter what they may be watching–whenever a particular show or commercial is aired. For example, suppose a company has created an ad that will run for three days. The first day, 25 percent of a potential audience saw the ad. On day two, 15 percent of the audience saw the ad. Finally, on day three, 20 percent saw the ad. The GRP for the ad would be 60 (25+20+15).
- Average Quarter Hour (AQH) takes note of how many households or persons on average watched TV for a minimum amount of time (typically five minutes or more) during a specific quarter hour of viewing (for example, between 8:00 and 8:15).
Diaries and Sweeps
Originally, in the early 1950s, the Nielsen TV ratings were drawn from a simple pen-and-paper diary. Each member of a Nielsen family was given their own diary and asked to write down what they watched. At the time, it was technologically impossible for Nielsen to keep a consistent day-to-day record of viewing, so the population sample was accompanied by a “sample” of the TV programming calendar.
The weekly diaries were mailed out and collected over a period of about one month, known as “sweeps,” which became characterized by TV channels offering their most compelling and sensational programming. Most markets could count on sweeps during four months (February, May, July, and November), but some markets had additional sweeps during January, March, and/or October.
The TV industry maintains a habit of creating programming schedules around sweeps, and the diary system still forms part of the data that Nielsen collects. However, the technological necessity of sweeps is mostly obsolete, and diaries have two obvious drawbacks: for one thing, you can write down anything you like, and for another, Nielsen and everyone else had to wait a long time for the data.
For more accurate and timely data than diaries could provide, Nielsen developed a system of electronic viewing meters and a reporting device. The individual meters are installed in or on the various TVs (and connected devices) in a home and networked with a “home unit” that uploads the viewing data to Nielsen on a nightly basis. Over the years, the meters have become more sophisticated and accurate (and less intrusive), but by themselves, they can’t tell Nielsen who is watching TV.
The diary method was more specific about who was watching any given show but lacked the concrete electronic verification and timeliness of the metered technology. So Nielsen developed the “People Meter,” a simple box with a button for each member of the household to press when they were watching TV.
Unfortunately, People Meters can provide inaccurate data. Many people forget or grow tired of pressing their buttons each time (or simply grow tired and fall asleep while watching). And there’s no way of knowing if the right person is even pressing the right button in the first place. You can meter equipment, but not people.
Nielsen, the TV producers, and advertisers alike heavily invested in live viewing, but over the course of the past few decades, “timeshifted” viewing was adopted by a significant portion of the population.
In the late 1980s, Nielsen adapted its electronic systems to include VCRs, but while the devices showed when a specific channel was being recorded, it did not show what was being played back, an –issue that became critical in the mid-2000s when DVRs became the choice of a quarter of the TV watching population.
Luckily, DVRs are much better than VCRs at generating useful viewing data; TiVo, along with most cable and satellite providers’ set-top boxes, collects information regardless of whether the household is a Nielsen family.
To account for the timeshifting element, Nielsen introduced the “LivePlus” ratings, which include shows that are watched within a certain timeframe of their air date–there’s a “LivePlus Same Day” for people who watch a show within 24 hours, as well as “LivePlus 2,” “LivePlus 3,” and “LivePlus 7” ratings for viewers who watch a show later in the week.
More importantly for the crucial advertising side of the TV industry, Nielsen also developed the “C3 rating”. This doesn’t measure the show that’s being watched, only viewership during specific times when commercials are aired. As with “LivePlus 3,” the number refers to time-shifted viewing, which can occur within three days of the original airtime.
The Virtual Challenge
Nielsen ratings now face a far bigger challenge than timeshifting. The ubiquity of online and mobile usage has provided options for TV viewing that make metering a DVR seem easy by comparison. These options include streaming services such as Netflix and Hulu, TV apps on smartphones and tablets, pirated downloads of shows, and an entire generation of viewers who are likely to “cut the cord” and not watch TV in any traditional way. All of these are modern ways to watch TV, especially among the crucially important young and affluent demographics.
In recent years, measuring the hard statistical data of the audience has become almost secondary to analyzing “engagement,” or the amount of online activity surrounding a given show. Nielsen responded by forming a social-focused division called NM Incite, which recently partnered with the highly-regarded SocialGuide to provide the “Nielsen Twitter TV Rating.” Social media provides a level of insight into viewership that mere channel viewing data cannot. In other words, it may be far more important and informative to analyze the “buzz” on Twitter and Facebook than to know whether a show is being viewed on the household television.
In a way, identifying the viewers and their demographic characteristics was more difficult at the time when the living room family TV set was the norm. TV viewing on a personal mobile device provides a very direct picture of the viewer; especially when that information can be cross-referenced with other publicly known data such as interests, “likes”, friends, and other social media information.
Still, there is difficulty in providing closer statistical correlations between social activity and viewership; especially advertising dollars. The relationship between advertising and social media or mobile apps is still far from an exact science. Then there is the further challenge of analyzing the usage of the TV services in which they are embedded (whether “second-screen” alongside traditional TV viewing, or alternate “first screen” methods such as streaming).
For decades, Nielsen enjoyed a virtual monopoly in TV audience measurement. Programmers and advertisers alike depended upon Nielsen ratings, and competitors provided only minor challenges to that three-way relationship. However, TV producers and advertisers have always looked for ways to supplement or replace Nielsen, especially when the competitors showed clear discrepancies between the ratings and their own analysis.
Nielsen’s technological edge has been threatened by the aforementioned ability of the TV providers and DVR services to collect viewing information from their devices–programmers and advertisers can get data directly from TiVo or Time Warner, for instance. And companies like Bluefin and Trendrr, who offer social TV analysis, make Nielsen’s partnership with SocialGuide less of an innovation than a necessary adaptation.
Ultimately, Nielsen has attained such a crucial position in such a widespread industry that it may be difficult to consider a world without Nielsen families and TV ratings. However, the TV industry itself has had to consider a world without TV–at least TV as we’ve known it for more than a half-century.