January 25, 2023
Commentary
Traditional TV Time Sinks During Higher Pandemic Viewing
- by Wayne Friedman , Staff Writer, April 12, 2021
Broadcast and cable TV erosion is accelerating -- at least via traditional measures. Nielsen’s prime-time broadcast viewing is down 20% to 24% or more in the fourth and first quarter, looking at the live program/same-day rating.
We expect double-digit percentage erosion declines. But a year before, they were in the 12% to 15% range. One possible answer, at least since October is that there are also far fewer prime-time original shows than a year ago.
Another factor is that Nielsen has been lax in serving its TV panel homes, due to pandemic issues that prevented its field agents from getting into homes. That’s important, research analysts say, in ensuring equipment and residents are using that equipment properly.
advertisement
advertisement
Nielsen says while it is true field agents haven’t had in-home access, they have been doing other things remotely. In addition it started in-home service again this past March.
And most importantly, it says ratings, demographics and other measures really haven’t changed. So no foul, right?
Brad Adgate, veteran media agency research executive and now a media consultant, says: “There could be an issue of fatigue.” In particular, this could include a range of things.
While a TV panel home may have the TV set on, people in the home aren’t always correctly pressing Nielsen People Meter remote controls. This lets Nielsen know who is actually watching, which then results in tabulation of demographic audience data.
Still, last week, the VAB blasted Nielsen for its relative lack of attention to these homes. One indication: The number of homes in Nielsen’s panel not watching “TV of any type.” This has grown to 4,642 in February 2021 from 2,735 in February 2020.
What’s going on here? Broken TV sets? Too much work outside the home? Weather-related incidents such as tornadoes, hurricanes and snowstorms?
Data can be vague. But opinions can be specific and pointed.
For example, there were 17 million Nielsen-measured average viewers watching the NCAA’s March Madness championship game last Monday night on CBS. Two years ago, there were 19.6 million for the big game.
Did 2.6 million go to streaming platforms to watch the game? (One major streaming platform distribution would like you to believe so.) Maybe time was spent watching more TV reruns of long-gone crime procedural dramas. Maybe those viewers turned to political podcasts.
Down the line, we can imagine a more precise postmortem will be revealed, explaining what really happened in the pandemic.
4 comments about "Traditional TV Time Sinks During Higher Pandemic Viewing".
Read more comments >
Next story loading 
Wayne Friedman is West Coast Editor of MediaPost. You can reach Wayne at wayne@mediapost.com.
Wayne, most o fthe national TV timefor the third quarter of 2020 was purchased during the 2019-2020 upfront and was based on the old rating levels then prevailing. So if viewing rose significantly because of the pandemic---as we all believe---this would have been a bonus---or unexpected --audience dividend for most upfront advertisers. Only scatter buys would have been distorted by slightly lower---not higher--- ratings during this period.
Looking to the future, once we get the pandemic under control it's most unlikely that future "linear TV" ratings would stayed at artificially elevated levels. Far more likely is the continuation of the steady shift of viewing time from "linear"" to streaming. So,when it comes to predicting ratings for the 2021-22 upfront, for example, Nielsen's 2020 lower-than-real ratings, would not be so far off the mark as the basis for negotiations---assuming that the demos weren't radically altered.
Panel attrition can conceivably drive reported tuning down, especially to the extent that households with more sets (and thus more hardware) are more likely to fail in-tab, and are also likely to skew to heavy TV viewership. Even then though, if Nielsen weights by number of sets, this phenomenon could be mitigated to some extent.
A couple of things happened during and as a result of the pandemic. First, cable and satellite households continued to decline, with the hit on the economy contriobuting to some consumers choosing to cord-cut or cord-shave for econominc reasons. Second, the inexorable shift in share of tuning from linear to streaming accelerated; if you're an avid Facebook user, then you are probably used to people in your feed looking for the next hot shows to binge.
As regards audience measurement, I've come to two pandemic-related conclusions. One, we can't rely on panels that require high-touch, in-home servicing, becsuse I have a bad feeling this won't be our last pandemic. And two, audience measurement needs to evolve to be fully platform-agnostic, so we can track both content and campaigns on both linear and streaming, and to integrate the two.
I think that we all recognize the problem with panels. Their cooperation rates are very low and when electronic measurements for every setin the home must provide data, day after day, for two or three years in a row--or longer, in some cases----this is a huge maintenance burden. We should also note that no amount of statistical sample balancing will ensure that the findings are representative of the nation as a whole if there are hidden biases in the recruitment ---and replacement---- process. For example, are heavy viewers of all ages more likely to agree to have the meters installed---because they are more favorably inclined towards TV? Is this question even dealt with---there are ways---in the recruitment/replacement process?
The whole idea underlying the peoplemeter system---as originally developed by Audits Of Great Britain many years ago ---was to combine meter set usage with an electronic diary surrogate to obtain better viewers-per-set data than pure diary measurements offered. Nobody even dreamt of obtaining minute by minute---or second by second viewing information---especially for commercials. Yet pressured by the agencies, Nielsen was coerced to produce "average commercial minute Viewing" data which is clearly highly inflationary and misleading to advertisers. That's on the agencies, not Nielsen. They demanded something which the peoplemeter system was never designed to do. But they got it.
Continuing my reply---- What's the most practical solution? It is often proposed that a massive panel---much larger than Nielsen's---be organized and, somehow, the usage of all devices with screens would be calibrated to generate household TV audience projections on a "granular" basis. Fine---but how do you recruit and maintain such a huge panel? Who pays for it? Who checks to see whether every device a home is using is being tracked on a continuous basis? What happens when a person moves to a new town or abode or buys a new device and starts to use it? And, most important? how do we get viewing data for each household member---not device use but viewing?What do we do about unattended device activity? In short, we would be faced with many of the problems that now cause concern about Nielsen's "outmoded" rating system. Is it really outmoded or is the problem that we are trying to obtain too much information in overly fine detail but lack the methodology for obtaining it? Arewe becoming too reliant on "data"?
I don't have a good answer to most of these questions but one thing I would recommend is that we start to think about what information we absolutely must have and what else would be useful and prioritize them. Next, we need to evaluate the various methods for obtaining said information to see what is possible and what is not. Finally, we need to broaden the discussion so that all parties who use the ratings are involved in a hands-on manner---and this includes significant participation---and funding---- not only by the media time sellers but also by advertisers, their agencies and program producers. If the time sellers pony up most of the money you are putting yourselves in the same boat that Nielsen now sails in.