Co-viewing On CTV Apps Slightly Outpaces Linear TV

Connected TV (CTV) does better than linear TV when it comes to co-viewing, according to new research from TVision, a TV measurement company.

Overall, CTV averages 1.29 viewers per viewing household (VPVH) with linear TV slightly lower at 1.26.

These results come from a study conducted from June-December 2021, looking at co-viewing time spent.

For marketers, co-viewing remains of high value -- especially where families with both kids and adults are watching at the same time, and messaging can be extended across many demographics.

Among ad-supported connected TV platforms, Tubi earned the best results of an ad-supported connected TV app for co-viewing, with around a 1.6 VPVH.

HBO Max was next at just over 1.4 VPVH -- followed by Hulu, also at 1.4, YouTube with 1.37, and Discovery+ at 1.36 VPVH.



Movies had the highest co-viewing score, at 73.2% of the time, while TV episodes came in at 54.8%.

The five best individual TV series are: Paramount+’s “The SpongeBob Movie: Sponge on the Run” (2.2); HBO Max’s “Esme & Roy” (2.1); Discovery+ “Battlebots: Bounty Hunters” (2.1); Peacock’s “Boss Baby” (2.0); and Paramount+’s “Paw Patrol: The Movie” (2.0).

According to the research, the best time for co-viewing is 8 p.m.

Younger audiences under 18 years old and 18-24 viewers are more likely to be co-viewers -- at 69.9% and 63.9%, respectively.

Who is watching the advertising while co-viewing? Adult audiences pay more attention to ads overall when they are watching with other adults -- with a 107.6 "attention index." When adults are watching with kids, they register a 100.1 index.

Children have less attentiveness overall. When watching with families, including adults, children have a 96.0 attentive index, compared to 97.1 when watching with other kids.

Domino’s had the most time spent co-viewing for a particular pizza brand during the survey -- at 39.8% -- followed by Little Caesars’ at 12.4%, Papa’s John’s with 11.4%, and Pizza Hut at 10.4%.

TVision used cutting-edge “computer vision technology” -- so called eyes-on-the-screen technology -- to measure person-level, second-by-second TV engagement. It counts every second of programming and advertising on television.

Every time a person walks into the room, its technology detects who the viewer is, where they are in the room, and what their eyes are looking at.

TVision’s data is collected from an opt-in panel of 5,000 homes across the U.S. Data are weighted to represent the country.

All demographic data is self-reported by the respondents.

TVision defines the co-viewing rate as the measure of episode or app viewed that occurs with another viewer present for five or more minutes.

The attention index is the average amount of seconds that viewers engage with a program.

Viewers Per Viewing Household (VPVH) is the average number of viewers present in the home when the TV is on with the content tuned.

3 comments about "Co-viewing On CTV Apps Slightly Outpaces Linear TV".
Check to receive email when comments are posted.
  1. Ed Papazian from Media Dynamics Inc, April 1, 2022 at 10:19 a.m.

    Wayne, Tvision has also reported that CTV commercials are about 15% lower than "linear TV" commercials in its averages for ad attentiveness levels. That's quite a different picture than what is suggested by the fairly small coviewing differentials---in CTV's favor.

  2. Jon Mandel from Dogsled Enterprises Inc replied, February 9, 2023 at 6:27 p.m.

    Ed, with all due respect, Tvision claims to be a "representative sample" representative of normal people is it if you allow them to watch you watching the telly? What about the bedroom tv? What about co-viewing then. Just because you got the age/sex/ ethnicity etc etc right doesn't mean you got it representative. This whole thing is just another case of the industry confusing and chasing bad kpis. We are both old enough to remember garbage in garbage out.

  3. Ed Papazian from Media Dynamics Inc, February 9, 2023 at 7:23 p.m.

    Jon, I have been into attentiveness research ever since the 1960 and I have not only seen just about everything that is publically available, but I have also done surveys of my own and initiated others. Regarding TVision having a nationally representative panel, I doubt that this is the case with the kind of perfection that  we might desire and until recently, TVision was  using its webcams only for one set---the most used one---in each home. I understand that this sitution may be  in review or is being corrected---but I don't know for sure.

    Anyway, returning to the question at hand, I am familiar with all sorts of surveys on this subject and I can testify that generally speaking, most of what TVision has been reporting is confirmed  directionally by the other research. What I can't say---and nobody really can say---is that a TVision finding that 37.5% of the program viewers for a low rated series watched a Coca Cola commercial while only 35.9% of the same folks watched a commercial for Tide and take that as a meaningful difference.That's cutting it far too fine for a panel with only 5,000 homes. But I can attest to the likiehood that both commercils were probably seen by about 35-40% of the people  in the room just prior to the break---not 93-94% which is what the people meters are telling us.

    A while back, I proposed the idea that the"ideal" national TV rating service should have a rating panel of about 150,000 -250,000  fully metered homes and that a sub sample of the larger panel-- say, 75,000 homes---maybe more---should include an attentiveness measurement of the TVision type. This would give us a sample large enough to generate fairly stable attentiveness factors for most national TV buys---if not for each episode of each series, then certainly for one network's bundle versus that of a rival seller. I haven't changed my mind about that.

Next story loading loading..