Even those who pay very little attention to the television industry have at least a vague notion that “ratings” are important. High is good. Higher is better. The shows with the best ratings attract the best advertisers, who in turn pay the most money for exposure to the large audiences those high ratings represent. It’s worked pretty much this way for years.

Until it stopped working. In September 2021, when Nielsen lost its accreditation from the Media Ratings Council (MRC), all hell broke loose, with broadcasters and advertisers forced to cast about for alternatives that might keep the television advertising machine running. Nielsen claimed the suspension would last just a few months, but in November 2022 the MRC’s audit committee announced that it was leaving Nielsen’s suspension in place—in spite of “significant progress on most of the issues that led to that suspension.”

Many marketers see this moment as a chance to reset an imperfect system. And pretty much everyone is puzzling over how best to move forward. We have some thoughts.

How Did This Happen? 

We could sense something was afoot in early 2021 when Nielsen announced that it would be adding broadband-only households to its local television metrics. Advertisers thought this was a stupendously ridiculous change in the metrics inasmuch as cord-cutters do not typically have access to local stations. After significant pushback from its clients, Nielsen finally capitulated (somewhat), agreeing to delay the change and to exclude from its new metrics any broadband-only household without local access. Advertisers still didn’t like how the shift would throw off their historical benchmarks, but OK fine.

However, even before Nielsen could implement that modified change, MRC announced that it had found a “generally consistent pattern of underreporting of viewing” by Nielsen. And not by just a little. MRC found that during February 2021, total TV-viewing by adults 18 to 49 may have been understated by anywhere from 2% to 6%, resulting in an undercount worth as much as $234 million to the networks. Multiply that by 12 and you’re looking at a potential annual shortfall worth nearly $3 billion. Given the complexities of upfronts and makegoods and other machinations associated with broadcast television accounting, it’s a mess.

Blame the Pandemic . . . But Not Just the Pandemic

We can’t say for sure, but TV execs think the root of Nielsen’s measurement screw-up can be traced to adjustments made due to the Covid-19 pandemic. For one thing, during the lockdown Nielsen stopped sending out field agents whose job it is to check in with empaneled “Nielsen families” to ensure the accuracy of its viewer tracking. In addition, some believe that lack of maintenance during the pandemic may also have contributed to the undercount.

Even so, it's not like Nielsen was beyond reproach prior to the pandemic. For years, marketers have complained about the firm’s monopoly and its various limitations: its flawed methodology, its inability to measure advanced platforms and cross-screen media, its troublingly small panel, its high price (for ad sellers especially). And it’s not like this was the first time Nielsen has been accused of under-counting over the years. In that sense, when MRC withdrew accreditation, it triggered a broader conversation about the shortcomings that have always inhibited our ability to measure television viewership and performance with a high degree of confidence.

Big Changes Are Imminent

In addition to fixing the issues identified in the MRC audit, Nielsen has pledged to make further modifications in response to its critics. Of particular note, Nielsen’s new approach promises to expand beyond the panel-only methodology of the past to incorporate big data elements as well—a much-needed shift in the era of smart TVs.

The networks love this promised adjustment. Because so many younger TV viewers are rejecting linear television in favor of streaming alternatives, it has become increasingly difficult for networks to deliver promised impressions among the most popular demographic cohorts (adults 18 to 49 primarily). Consequently, publishers are pushing to base their guarantees on total audience delivery instead, with narrower targeting possible (at a significant premium) only through their big data solutions. In essence, they’re saying: If you want more precise targeting, you can have it; but you’ll have to pay for it.

So far, advertisers have successfully pushed back against this total-audience gambit, but it seems unlikely that they will be able to resist forever. For now, the networks are insisting that the conversion to total-audience measurement and delivery will be implemented as part of the ‘24/’25 upfronts. Also looming on the horizon: a push to move away from impressions to “impact ratings”—but that might be a topic for another time.

Not the Only Game in Town

Whether Nielsen remains a part of this new measurement solution still remains to be seen. While it has been working more or less around the clock on fixing and improving its television measurement services, up-and-coming competitors have been slowly gaining favor within the advertising industry. Just a month after Nielsen lost its accreditation, NBC created a forum for industry leaders to consider alternative approaches to television audience measurement. Many prominent advertisers and agencies (including Canvas) participated in the forum, and more than 80 potential measurement partners responded to the resulting RFP. If you’re Nielsen, that’s an enormous competitive set.

At this point we should note that, in the end, we may not be talking about a single winner but rather about several winners in the race to become the new currency for television measurement, as many now question whether a single-currency solution is even feasible in the age of platform proliferation. About all we can say for sure at this point is that, regardless of how it all shakes out, TV “ratings” will never be the same.

So, What Are We Doing About All This? 

(This part could get a little nerdy, but we share it without apology. We are media nerds, after all.)

Beginning in the Spring of 2022, Canvas engaged in discussions with six top measurement companies (Nielsen One, Comscore, iSpot, VideoAmp, Samba TV, and 605) in order to identify the most seamless, scalable, and interoperable TV measurement solution available. We asked each firm to respond to an RFI based on the following evaluation framework:

1. Ability to measure all viewing sources end to end (25%)

Linear and digital viewing sources, ad occurrence methodology, conversion, people panel, estimates, methodology, and weighting

2. Methodology for collecting and counting data (15%)

Impression definition, completion rate, deduplication, and measurement identity

3. Quality of viewable impressions (30%)

Length of creative impression required to “count”

4. Reporting and optimization (15%)

Granularity of reporting metrics and optimization capabilities

5. Scale and viability (10%)

Type of vendor, geographical coverage, and cost

6. MRC accreditation (5%)

Extra credit

After a series of partner deep-dives during the Summer of 2022, we performed side-by-side comparisons to narrow our list based on a common scorecard, with particular focus on data stability. As you might guess, there was fallout: We chose to eliminate both Samba TV and 605 due to a lack of scale and capability in comparison to the other contenders. This initial evaluation further confirmed that, although Nielsen still lacks accreditation, its solution is nonetheless on par with (if not somewhat better than) its primary competitors’.

The Final Four

That leaves us, then, with four measurement partners still under consideration as we search for an accurate measure of “true” television ratings: Nielsen One, Comscore, iSpot, and VideoAmp. The final selection, of course, will come down to performance.

Consequently, in December we began a series of in-market tests, with a first round of testing deployed in December and further tests continuing through Q1 2023. During this final assessment, we will be checking for missed or overdelivered impressions and running side-by-side comparisons evaluating the stability of each partner’s data compared to both current and past schedules. (We’re trying to understand the impact on delivery and the corresponding cost implications.)

As a final pair of considerations prior to making our ultimate recommendation, we’ll evaluate each partner on two critical factors: 1) their ability to help us in our stewardship to ensure that each network delivers on its audience guarantees; and 2) the marketplace consensus on who is emerging as the best-in-class solution. In order to ensure that we have a seat at the table as that industry consensus emerges, Canvas remains active in several measurement councils, including NBC, IAB, and the 4A’s.

In Search of Consensus

Consensus could prove elusive, however, especially if the best option requires more than one measurement partner—a so-called “multi-currency” solution. Ideal though it may be, there could be so many hurdles that going with multiple currencies just isn’t viable. Here are just a few of the factors that could get in the way:

• Higher Service Costs – Nielsen is pricey enough as it is. Having to subscribe to multiple measurement systems may simply be cost-prohibitive for most.
• Price Increases – Multiple currencies will require investment negotiations using completely different bases and methodologies, which ultimately will work to the advantage of the sellers (the networks).
• History – Most negotiations are built on historical models which will no longer work. Establishing new benchmarks for negotiations and reporting will take time.
• Interoperability – Even if it’s the best solution, will networks and agencies have the ability to manage the complexity of a multi-currency approach?

As you have probably guessed, a final, elegant, industry-wide resolution to all of this is still months away. But that does not mean we aren’t close to knowing what needs to be done. Although the full solution will not be “ready for primetime,” we expect to have a recommendation for our clients in time to begin our upfront negotiations in the Spring. (Full implementation of the new approach is still likely to be at least a year away, however.) In the meantime, if any of you nerds are interested in seeing our scorecards and evaluations so far, we will be happy to set up time to take you through the math. (We media nerds need to stick together, after all.) Needless to say, there is still a lot of work to do as we search for the best way to measure television ratings going forward. As they say in the TV business, stay tuned.

---