Audience Insights: Local TV’s Perplexing Population Puzzle

To paraphrase a 1984 Depeche Mode classic, “People are people, so why should it be…you and I should [watch TV so differently]?”

Some of the answers are straightforward and obvious, based on our interests. You may not be a sports fan. I, on the other hand, still occasionally check the program guide in hopes that ESPN Classic has made a comeback. It could be that you appreciate a good drama, while I prefer comedies. Perhaps you’re firmly entrenched in the 18-49 demographic, while I may (or may not) be clinging to the upper end of the 25-54 bracket.

While those and other differences understandably drive variance in viewership levels across TV networks, dayparts, and programs, at least we can agree on how many of us are out there in the population – or so it would seem.

Interestingly, though, continuing changes in how we access and consume television content are creating challenges when it comes to counting how many potential viewers are out there – much less how many are tuning in to your favorite Cable TV network on a frigid Friday evening in the fall or sunny Saturday afternoon during the summer.

One persistent issue facing measurement companies is that population counts, by definition, are moving targets. For evidence, look no further than the Census Bureau’s Population Clock, which tells us that the U.S. population had a net gain of four people in the two minutes it took me to pull up the site and embed the hyperlink. This constant change means that analysis of what’s happening among the “general population” (or subsets thereof) – including how and when we watch TV – is rooted in snapshots, estimates, and projections of how many of us are out there to begin with.

And those numbers don’t always align perfectly. To illustrate, let’s look at key population figures used by the major TV measurement companies, Comscore and Nielsen. For benchmarking, the U.S. Census Bureau’s “QuickFacts” page shows an estimated 122.4 million total households throughout the country, based on American Community Survey (ACS) estimates covering the five-year period of 2016-2020. Heading into the 2022 TV season, TV ratings stalwart Nielsen projected that the total number of U.S. households had topped 127 million, with more than 122.3 million of those qualifying as TV households. But Comscore sees the landscape a bit differently, coming in at a comparably lower estimate of 95.4 million U.S. TV households.

While proprietary modeling methodologies can drive some variance, much of the gulf between 122.3 million and 95.4 million comes down to three letters. BBO is the industry acronym for broadband-only homes, which are TV households that rely solely on Internet-based streaming services to get programming to their sets. For additional clarity, BBO homes do not subscribe to Cable or Satellite services and do not receive local Broadcast stations via antenna. When BBO is factored out of Nielsen’s 2022 estimates, approximately 99.7 million traditional TV homes remain. This more “apples-to-apples” comparison puts Nielsen and Comscore relatively close in terms of household population.

Moving beyond the household level, things get a bit more intriguing when we focus on the segment of the population falling in the 25-54 age bracket. Based on Cox Media analysis of Census data, these adults 25-54 (A25-54, for short), represent just over 39% of the U.S. population. Applying that to the QuickFacts population of 331.4 million would yield a 25-54 population of approximately 129.7 million.

At first glance, Comscore’s A25-54 population estimate (also commonly known as a universe estimate, or UE) jumps off the page at a surprisingly low 60.3 million – but there is a valid explanation that is important to keep in mind, especially for marketers who work with Comscore local TV data. Unlike Nielsen, Comscore does not attempt to project the number of people falling within certain age ranges. Instead, all of Comscore’s demographic data are based on households, which means 60.3 million is the company’s estimate of how many non-BBO U.S. TV homes contain at least one adult 25-54.

By contrast, Nielsen’s people-based A25-54 universe estimate for 2022 stands at approximately 121.6 million, which on the surface seems more in line with Census-based projections. That said, a closer look shows some notable differences. For starters, Cox Media calculated the estimated number of A25-54 per household, which sits at approximately 1.06 using Census data and just over 0.99 using Nielsen data. That disparity grows when BBO homes are isolated within Nielsen’s estimates. Based on Cox Media analysis, Nielsen projects only 0.86 people 25-54 for every traditional TV household, compared to a whopping 1.58 among BBO homes.

If your head is spinning a bit, you’re not alone – and you’re probably wondering why we’ve spent the last few minutes studying population statistics. Let’s wrap things up with three takeaways:

1. Measurement is not easy.

If gaining consensus around Census stats is challenging, so too is finding agreement and absolutes when trying to count local TV viewers every hour of every day.

2. Cox Media offers choice for TV advertisers.

For local TV measurement services, we continue to partner with and gain perspective from both Comscore and Nielsen.

3. Cox Media delivers advanced analytics.

Besides estimating the overall viewership for local TV campaigns, your Cox Media team can provide audience-based planning beyond age-and-gender demographics, as well as insights on audience engagement and campaign attribution.

Add it all up, and your local Cox Media team is ready to partner with you now to navigate all the numbers and nuances of your next successful marketing campaign.

Follow us here for more Audience Insights updates, and feel free to contact Cox Media with questions or topic suggestions for future posts.