When it comes to national television measurement used for the buying and selling of commercial time, Nielsen is an industry-mandated monopoly. While potential competitors have periodically emerged, they failed to sustain enough support to survive – not necessarily because Nielsen did anything to impede them, but because advertisers and networks have been unwilling to pay for multiple sources to provide similar national TV audience data.
While there have been occasional calls for change, the industry’s unwillingness to support any real competition to Nielsen, along with networks and media agencies continuing to sign long-term Nielsen contracts, make the status quo is hard to budge.
Why Alternatives to Nielsen Have Failed to Sustain Support
There have been numerous unsuccessful attempts by other companies to compete with Nielsen, both on a national and local level. The most common result has been occasionally pressuring Nielsen to make improvements to its own methodology and increase its sample size.
Here’s a brief look at the most notable attempts to compete with Nielsen.
- 1985: After more than 30 years of being essentially unchallenged in the national TV audience measurement arena, British company AGB entered the game and posed the first real threat to Nielsen’s dominance. It had been using something called people meters in Europe, and saw an opportunity to expand into the U.S. market. AGB had the advantage of actually using a better methodology than Nielsen’s antiquated meter/diary approach.
AGB spent nearly two years testing its people meter in Boston before trying to roll it out nationally. Nielsen responded by launching its own people meter, and increased its sample from 1,200 to 2,000 homes (to match the size of AGB’s sample). The following year Nielsen increased its sample to 4,000. As a result, AGB failed to sustain industry support and ceased its U.S. operation in 1988. The main problem was that ad agencies were not willing to pay for two identical national measurement services, and most of the networks were not willing to support a new ratings system that did not report higher numbers than Nielsen (I believe at the time only CBS and MTV were willing to go forward).
- 1987: D. Percy & Co. developed a passive meter, which used a heat-sensitive, infrared device to detect when someone was in front of the television set and when someone left the room. But when people found out that the heat sensors could be thrown off by the presence of a large dog, jokes ensued, and any potential support fizzled. Nielsen announced its own plans to develop a passive meter, which never went beyond the testing phase.
- 1991: Arbitron’s ScanAmerica was way ahead of its time as it tried to marry TV viewing to product usage (sample participants were asked to run a “scanner wand” over the product code on each store-bought item). Single-source data was considered the holy grail at the time, and this approach seemed to be what advertisers had been seeking. But a lack of belief in the methodology, the limited number of “scannable” items, and an unwillingness to invest in an unproven system in a faltering economy, resulted in ScanAmerica shutting down only a year after being introduced. Had ScanAmerica come along a few years later, when the economy was back on track and other viewing sources were cutting into network ratings, it might have received more interest.
- 1993: Arbitron, which had been competing with Nielsen for four decades in collecting local TV ratings (in roughly 200 markets), canceled the service. Local stations, which had been forced to subscribe to both Nielsen and Arbitron to accommodate advertisers and their agencies, who were more or less equally divided among the two services, were under increasing economic pressure from cable and other competition. It no longer made financial sense for them to support two essentially identical rating systems. Nielsen, which already had a lock on national TV measurement, won the day. Nielsen acquired Arbitron (renamed Nielsen Audio) in 2013.
- 1993: The broadcast networks tried to develop a rival rating service, SMART (System for Measuring and Reporting Television), with a new people meter provided by Statistical Research, Inc. (SRI). But they couldn’t get the necessary ad-agency support and it never got beyond the “media lab” stage (I thought at the time SRI did great work and provided valuable insights). It folded in 1997.
- 2007: erinMedia’s plan to launch a new national ratings system to collect and process viewing data from set-top boxes was derailed when Nielsen announced it was forming a new unit called DigitalPlus to essentially do the same thing (which caused erinMedia’s funding to disappear). In 2005, erinMedia had filed an anti-trust lawsuit against Nielsen Media, which was confidentially settled in 2008.
- ComScore, which merged with Rentrak in 2016, stands as the only real competition to Nielsen – primarily with local TV stations. Most major media agencies, networks, station groups, and streaming services subscribe to some level of ComScore data.
So, when it comes to national television audience measurement, Nielsen remains a virtual monopoly, albeit an industry-mandated one. Nielsen ensures its monopoly status, with its clients’ approval, by continually signing long-term staggered contracts with networks and media agencies. This means any fledgling measurement company cannot actually replace Nielsen, but must operate side-by-side (at least for several years). Since the industry has long been loath to support two essentially identical rating services, particularly for national currency measurement, Nielsen has continued to operate largely unopposed.
When Nielsen Has Been Forced to Improve TV Measurement
There have be only two instances, more than 20 years apart, when Nielsen was forced to overhaul how it reported television audiences – in 1987, when the switch was made from the antiquated household meter/diary system to national people meters, and in 2009, when DVRs and time-shifted viewing led to the shift from program ratings to average commercial-minute ratings (C3).
- In 1985, when AGB, which had been people meters in the UK, tried to introduce them into the U.S. market, this represented the first time a real threat to Nielsen’s dominance emerged. Even Nielsen had to admit that this new electronic measurement system was far superior to its own meter/diary system, which had considerable human bias attached to it – the highest rated networks and programs, as well as shows airing multiple times per week (daytime, early morning, evening news, late night, syndication) benefited when people manually filled out diaries at the end of each week. Recall tends to favor the most popular and longer-duration programming. Nielsen was forced to develop its own people meter system.
- In 2007, commercial avoidance through fast-forwarding had become one of the biggest advertiser concerns. DVR penetration among U.S. TV homes hit 20%, and measuring DVR playback and delayed viewing was now a priority. Virtually every study and survey ever done indicated that anywhere between 70% and 80% of VCR or DVR playback involved fast-forwarding through commercials. But there was no actual Nielsen data to prove it, and Nielsen’s national people-meter sample had just started including DVR homes. Nielsen was forced to report DVR playback (something it was never able to do with VCRs), which led to C3 (average commercial minute ratings up to three days after the original broadcast).
How these two fundamental changes were implemented shows how much the industry’s priorities have changed over time – the speed of change in the access to video content and in consumer behavior has caused good research to take a back seat to rushing new products to market.
There was a time when any changes to audience measurement needed to undergo extensive vetting before the industry would consider implementing them. People meters were evaluated by industry committees for nearly three years – including a full year of side-by-side data to compare it with the old meter/diary method – before agencies and networks decided to use it as marketplace currency.
Before switching from average program ratings to C3, however, little research was done into the methodology for measuring average commercial minutes. This was the first time that marketplace business concerns overrode the necessity for validating research methodology. It is also the moment that Nielsen shifted away from being a pure research company – at the following year’s national client meeting, Nielsen actually proclaimed that getting a product to market was now more important than validating the research beforehand.
Does Anyone Really Want More Accurate TV Audience Measurement?
The major impediment to improved audience measurement is that despite claims to the contrary, neither buyers nor sellers have traditionally wanted more accurate audience measurement. Sellers want higher ratings (so they can charge more for advertising) and buyers want stability (so they can project future performance). More accurate audience data is conducive to neither.
- Sellers want higher ratings – you never hear any network complain that Nielsen is overcounting their audience (although for years it did). But every time I can recall reported television usage or network ratings declining suddenly, or for no reason anyone can pinpoint, they blame Nielsen and call for change. There’s a reason some networks pushed so hard for viewing in bars and restaurants to be included in currency measurement? Anyone who knows anything about viewing habits or audience measurement knows how absurd it is to think more than a small fraction of people in bars or restaurants watch and hear commercials. It’s all about higher reported ratings. When Nielsen was unable to report VCR playback, on the other hand, the networks were happy to continue using the artificially inflated numbers Nielsen reported (for about 20 years, until DVRs overrode the need to properly measure VCRs).
Any new measurement system does not provide higher broadcast network ratings than Nielsen, was not able to sustain major media company support. Today, however, with the major media companies owning linear and streaming platforms, losing broadcast viewers but gaining more viewers elsewhere may be more palatable.
- Buyers want stability – accuracy is fine, but stability is required to be able to project future performance. Anyone who was in a media agency research group when the industry switched over to buying and selling based on C3 remembers the chaos that ensued during the upfront and for much of that year. There’s a reason that what was supposed to be a “one-year band-aid” is still being used 13 years later. Moving to more accurate commercial measurement would require not only additional staff, but would result in significantly less stable and projectable ratings.
Back in the days when there were relatively few viewing options, and most networks started and ended their seasons at the same time, there actually was a degree of stability to people’s viewing habits. Not so today. Traditional TV seasons, with new series debuting in the Fall and Spring, might still exist for the broadcast networks, but not for other platforms (or for viewers). Streaming services drop new seasons and new shows throughout the year, whenever they are ready, and not always at the same time as previous seasons.
Some streamers drop entire seasons at once, some two or three episodes before switching to weekly, and others use the broadcast network once-a-week model. When a new season of Stranger Things, or a new show like Squid Games or The Recruit premieres, for example, my viewing habits immediately change. I’ll binge a few episodes a night, and DVR my regular linear shows (which I’ll binge two or three weeks later). My viewing is not nearly as stable or predictable as it was 10 years ago, and much of my viewing would not be captured by Nielsen’s C3 (or C7) measurement.
- And let’s not forget that agencies do not want to incur dramatically increased costs for improved ratings systems, or multiple currencies that would require significant spending for new or custom databases (and staff) without increasing their revenue.
Can Samples Still Adequately Measure the TV Universe?
The whole purpose of a sample is to be representative of the total U.S., with the idea that you and your demographic cohorts have similar viewing habits. Your cohorts are determined based on characteristics that Nielsen has determined impact viewing behavior – such as age, sex, race/ethnicity, presence of children, income, education, language spoken at home, etc. Also important is whether or not you have cable/satellite/telco (i.e., pay TV services) or DVRs.
In the past, broad platform mattered more than individual platform. In other words, whether or not you have cable does not require the sample being representative of every cable system, because that does not significantly affect viewing choices or behavior. Nor is it important to know what type of DVR you have, since that will not substantially impact your delayed viewing, or how much you fast-forward through commercials.
Whether or not you subscribe to streaming services, however, is not the same. If you only have Netflix, the programming available to you, and your actual viewing habits, will be dramatically different than if you only subscribe to Paramount+. Likewise, if you bundle Disney+, Hulu, and ESPN+, and add HBO Max, your viewing habits will be different than if you have Netflix, Prime Video, Peacock, and Apple TV+. And there won’t just be differences in what you stream, there will be significant differences in your linear TV viewing as well. The only way to have a meaningful television sample, is to accurately represent the total U.S. based on every possible combination of streaming service subscriptions and cable/satellite/telco. This means that the make-up of Nielsen’s national and local samples could need to change dramatically every year or two. The days when all Nielsen needed to worry about was cable and DVR penetration are gone forever.
But even if you are able to put together a sample that keeps up with the shifting combinations of platform usage, it won’t necessarily provide significantly more accurate data. This is primarily because as more and more niche programming becomes available on more and more platforms, it is quite likely that demographic cohorts no longer have similar viewing patterns.
Nielsen still can’t accurately measure streaming, and won’t be able to until its sample reflects individual streaming service penetration and until it can measure viewing on all platforms and devices. Netflix’s two minutes of viewing data is meaningless gibberish that tells us nothing about how many viewers are actually watching anything. The press should stop repeating (as opposed to reporting) these bogus numbers provided by streaming services. Is Squid Game really Netflix’s most popular series ever? Maybe. If ABC said 50 million people watched Grey’s Anatomy this month without any third-party verification, I doubt anyone would report it as a real number. In reality, no reported streaming data is anywhere near being comparable to linear ratings reported by Nielsen or ComScore.
The fact that Nielsen only reports on streaming titles from Netflix, Hulu, Prime Video, and Disney+ (it recently AppleTV+), and only measures streaming on a TV set, is an ongoing embarrassment. Netflix has repeatedly said that Nielsen’s limited reported streaming data is “not even close” to the actual numbers.
There are actually more high-quality shows on television than ever before. The most popular ones seem to be on streaming services. It seems that way because the press headlines, social media chatter, and industry “buzz” tell us so. Are Stranger Things or Squid Game really more popular than NCIS or Chicago Fire? Of course not. It’s just that we’re not used to evaluating linear TV shows based on time spent viewing or combining multiple sources of their viewing (because advertising isn’t bought and sold that way).
In reality, we have no idea how many people (or who) are watching any individual streaming show or platform. Whatever limited data is currently provided by streamers is not using the same metrics as Nielsen (or one another). So, we really do not have any valid ongoing comparisons between linear TV viewing and streaming using the same metrics or time frame. And any comparisons we do get compare minutes of total viewing among shows where some have eight episodes available to stream to others that have more than 100 episodes available. We do know that broadcast ratings tend to decline every season. Streaming series may as well, but we just don’t see the data.
Will New Vendors and a Joint Industry Committee Change Anything?
A number of advertisers, media agencies, and industry trade organizations joined NBCUniversal’s Measurement Innovation Forum, which is designed to evaluate current audience measurement methodology and come up with alternatives to Nielsen. NBCU sent out a request for proposal (RFP) to more than 50 companies (including Nielsen) to help “build a new measurement ecosystem for us that reflects the future.” Ironically, this was not in response to the basic flaws in current C3 measurement, but rather was spurred by the problems Nielsen encountered maintaining its national people meter sample during the pandemic – which resulted in reported network audience declines that seemed illogical to say the least. Again, media companies only seek action when they believe reported ratings are too low, not when they believe they are inaccurate.
Tests of potential alternatives to Nielsen (such as ComScore, iSpot.tv, Samba TV, VideoAmp, Xandr, and LiveRamp) have reportedly not yielded the results the industry is looking for. Different companies have been providing substantially different numbers from Nielsen and from one another. And even the same suppliers have sometimes produced widely different numbers from one week to the next. As I’ve already mentioned, in today’s media world, currently structured samples (and data modeling) are woefully inadequate in measuring the universe of TV viewers, so significant disparity in data produced by different samples should not surprise anyone.
If data provided by these different companies continue to dramatically diverge from one another, none is likely to be seen as a valid alternative to Nielsen as marketplace currency.
Obviously, a giant set-top sample in the range of tens of millions would be best, but it would still need to be representative of the country at large based on age, income, ethnicity, media device ownership, streaming subscriptions, etc. But even that would significantly under-report viewing sources with a large amount of over-the-air only homes, and would still need a Nielsen-type panel to ascribe demographic data to these homes. But if I had faith that the household sample was indeed representative of the country at large, I would have no problem with a panel for demos.
Now five major media companies (NBCUniversal, Paramount Global, Fox, TelevisaUnivision, and Warner Bros. Discovery) along with the Video Advertising Bureau (VAB) are forming a “joint industry” committee to create standards for audience measurement. Notably missing from this alliance (at the moment) is Disney (which owns ABC and ESPN as well as Disney+ and Hulu).
Since it can’t really be called a joint industry committee if it only consists of sellers, they recently announced that media agencies have been invited to join, or as they phrased it, “We believe that if we’re going to realize our collective ambitions of ensuring accurate cross-platform solutions for measuring premium video are available, we cannot do it alone. We want your feedback as we establish new standards for the future.” I find the use of the word “feedback” rather than “participation” interesting, to say the least.
There are just three times I can recall advertisers, media companies (sellers) and media agencies (buyers) in the same room discussing how to improve TV/video audience measurement. Even prior to the switchover to people meters, Nielsen held separate meetings among researchers from the networks and ad agencies – and neither the network salespeople or agency buyers were present.
- In 2005, the Council For Research Excellence (CRE) was formed (I was privileged to be one of its founding members). The CRE initially consisted of around 40 top industry researchers, representing advertisers, networks, local station groups, syndicators, and media agencies. While the CRE did conduct some landmark research, it did not lead to any substantial changes in Nielsen’s measurement methodology – not really surprising, since the “independent” Council’s research was funded by Nielsen (roughly $3.5 million per year). For those who don’t remember or weren’t around at the time, the CRE was formed following Senate hearings on Nielsen’s purported measurement bias, so Nielsen could avoid the federal government becoming involved in overseeing national TV audience measurement. Lawyers were present at many of our early meetings because the potential for collusion between buyers and sellers was a major concern. Nielsen pulled its funding in 2017.
- In 2009, television buyers, sellers, and researchers convened for a meeting with top Nielsen executives to discuss how to measure television viewing via DVRs. This was the first time top sales people and buyers were included in this type of meeting. Measuring delayed viewing had become a major concern among advertisers (Nielsen had just started including DVR homes in its national sample). There was urgency to arrive at a solution before the 2009 upfront. Advertisers and agencies, of course, only wanted live viewing reported as marketplace currency, while the networks wanted a full week of delayed viewing (C7) reported. C3 was the compromise.
- A few years later, the ANA convened a select group of advertisers, sellers, buyers, and researchers to discuss replacing Nielsen’s C3 measurement with true commercial ratings. This meeting ended with Nielsen, in its typical arrogance, basically telling us they heard our concerns, but had no intentions of shifting away from C3 anytime soon (this was 10 years ago).
Only extremely urgent marketplace forces and advertiser concerns have led to any real audience measurement improvement. Measuring streaming may qualify as one of those times – particularly with Netflix and Disney+ joining other streamers in offering an advertiser-supported tier. Although given that the major media companies own several of the major streamers (Disney+, HBO Max, Paramount+, Peacock, Hulu), I don’t see why they can’t provide combined actual viewing data in the same format to advertisers – unless, of course, they realize the data will not be as big as they are currently able to lead people to believe they are by releasing nonsensical data such as billions of minutes viewed.
What the Industry Needs to Do to Improve TV Audience Measurement
It’s interesting that even with all the criticism leveled at Nielsen, the company is still used as the base to compare any new potential vendors to gauge the validity their reported ratings. So, if another company reports data that is 30% higher or lower then Nielsen’s, it is immediately thought to be “wrong.” But we really have no idea if this is the case.
In addition to re-thinking sample construction, the first thing that needs to be done is to get a handle on what current rating services can and cannot accurately measure. While this might sound logical and simple, I can’t remember this ever being done (outside of the old telephone coincidental days). Even when the industry took the major step of moving to average commercial minute measurement (C3) in 2009, there was precious little analysis done into the methodology Nielsen uses to calculate these ratings.
The only good attempt to get some of these answers was in the CRE’s landmark Video Consumer Mapping Study in 2007-08. I was co-Chair of the CRE’s Media Consumption and Engagement Committee, which commissioned the study. One of the lesser-known findings was that Nielsen’s overall usage levels of households and broad demographic segments, such as total viewers and adults 18-49 were remarkably similar to the observed viewing behavior in our sample. Once you started to look at narrower age groups, however, the reported Nielsen data started to stray significantly from the observed viewing data. And this was just overall TV usage levels, not individual program ratings (and was 15 years ago, when video viewing was not nearly as splintered as it is today). DVR penetration was barely 20% of the U.S., and there were no streaming services. At the time, I suggested we replicate the study every five years, but that never happened – primarily because it was so expensive, and the CRE (and Nielsen) had other priorities.
It is actually not that difficult to figure out how accurate current TV measurement is. I bring this up every time I write on this topic. When I was on the CRE, I proposed a methodology for gauging how accurate Nielsen was at measuring an increasingly fragmented video viewing environment. Nielsen didn’t want to have anything to do with it then, but the industry should consider it now.
Here’s all you need to do:
- Develop a sample of 100 or so of the top researchers in the industry (buyers, sellers, advertisers), who are adept at detailed work and understand how audience measurement works. Nielsen, ComScore (and others) would meter their TVs, DVRs, smartphones, computers (desktops, laptops, tablets) and any other video viewing capable devices they own. They would also be given portable meters to measure their exposure to out-of-home video and audio. Let’s see how many actual people are actually watching the commercials in bars and restaurants compared to what the portable devices pick up as “exposure.”
- Set up several different viewing scenarios for what and how each participant will view video content and commercials during a single day (6am-1am), which would include both live viewing and viewing via DVRs (with normal fast-forwarding, pausing, etc.).
- Participants log in their video and commercial exposure and viewing (down to the second).
- Compare their actual viewing to what the meters report. We would then be able to see and address weaknesses in audience measurement, see how much real-world viewing is not being captured by current methodology, and gauge the differences between such things as exposure (the opportunity to view) and actual viewing, and between individual commercial viewing and average commercia minutes.
Not only will this provide, for the first time, a look at how accurate reported ratings are, but it will also tell us exactly where improvements to audience measurement need to be made (which was the original but unfulfilled purpose of the CRE). The project could be overseen by a group of retired research executives or some well-respected company with no skin in the game.
Until something like this is done, all this talk will be little more than the sellers trying to get higher ratings, or moving away from just measuring eyeballs to the even more difficult to measure behavioral and attentiveness data (which everyone has been discussing for years). Most attentiveness data I’ve seen over the years have been gibberish. There has always been a tendency to start focusing on the next shiny object before actually fixing what is still the biggest piece of the pie – in-home viewing to television sets. But if we don’t get that right, measuring real cross-platform viewing will remain out of reach.
So Whats Next?
That’s a good question. I doubt whether the industry will follow my proposal to discover the actual weaknesses in current audience measurement techniques. That would require the expensive, time-consuming process of fixing them, and not necessarily supply the answers anyone wants to deal with. So, if past is prologue, we’ll probably see more talk and several new proposals on how to “better” measure video and cross-platform viewing. New companies will come in with innovative techniques for measuring the video landscape (or portions of it), but won’t get the funding they need to go through the extended scrutiny the industry will require. Nielsen will be forced to make some changes to how it operates. When the next upfront rolls around, most everyone will likely still be using Nielsen as their primary national marketplace currency.
If the media conglomerates sign long-term contracts with Nielsen during the next round, we’ll know most of this was just talk. If they sign one-year deals, making Nielsen (and others) continually prove themselves, then maybe real change will happen.
Comments