Commentary
Time Shifted TV Viewing Is The Default
- by Jack Loechner , Staff Writer @mp_research, April 16, 2015
According to new research from Hub Entertainment Research “Time Shifting” viewers, who have broadband and watch at least 5 hours of TV per week, time-shift more TV than they watch live. The average viewer says that 47% of the TV shows they watch are live and 53% are time-shifted. Among Millennials, time shifting is even more common. Viewers 16-34 say that only 39% of the TV they watch in a typical week is live.
Most time-shifted viewing still happens through a set top box, says the report. Together, DVRs (34%) and VOD from a pay TV provider (19%) account for more than half of all time-shifted viewing. The study explores the usage and drivers of four main categories of time-shifting alternatives: DVR, Video on Demand, TV Everywhere and OTT platforms.
Untethering TV from a linear schedule has both invigorated and disrupted the TV industry, says the report. Watching on your own schedule has made TV a more compelling entertainment option for consumers, in general, and it has also made back catalogs more accessible, and thus more valuable. However, shifting has put tremendous pressure on how the TV business has traditionally made money: TV shows watched live, with ads
The executive summary overview of findings provides these details:
When it comes to TV viewing, consumers are increasingly living in the past
- 53% of all viewing among consumers16-74 is time shifted, vs. 47% live
- 42% prefer watching episodes of a current season via time-shifting, vs. 28% who prefer watching live
- Among Millennials, 61% of viewing is time-shifted
- 2.3 is the average number of different sources each viewer uses to time-shift.
- 53% of all time-shifting is done through a DVR (34%) or VOD (19%)
Ad avoidance is not the biggest reason people time shift—but it’s a factor
- 37% say the ability to skip ads is at least one benefit of time-shifted TV when fast-forward is available
- 41% say that having fast forward disabled on VOD is a major "frustration"
- 49% of VOD viewers fast-forward through EVERY commercial
- 56% skip every commercial when viewing from a DVR
In aggregate, DVRs and VOD service account for more than half of time shifted viewing:
Percentage Of Shows Viewed Per Source
- 34% Shows recorded on a DVR or TiVO
- 19% On-Demand service
- 16% Shows streamed from Netflix
- 8% Sites or apps from TV networks
- 6% Shows on Hulu or Hulu Plus
- 3% Amazon Prime or Amazon Instant
- 7% Any other online source
Top Reasons For Watching Shows Later (All Respondents)
|
Start with the DVR, sprinkle in VOD, and add OTT services, and suddenly the ability to watch TV on your own schedule has shifted from a benefit to an expectation, concludes the report. It's all part and parcel of the new mindset that the Internet has wrought: providers of any service must deliver in a way that meets the consumers' needs, rather than the other way around.
Time-shifted viewing is now the default, says the report. And, If anything, time-shifting is only going to increase. With the exception of sports and some reality genres, most TV shows today are watched sometime after they air. All of which makes the question of how to track viewership and monetize programming more challenging, and more urgent, than ever.
The complete report, with more charts and graphs, may be accessed here.
Jack Loechner is the Editor of The Center for Media Research's daily Research Brief. You can reach him at jack@mediapost.com.
I hope Nielsen is reading this. The percenage of viewing that is delayed, as estimated by respondents in this study-----is about 4-5 times higher than what Nielsen's meter panel is reporting. So who are we to trust----the mechanical and supposedly precise meters or the fallable and not very accurate human respondents?
I agree that delayed viewing is likely over-reported, but at least that tells us what respondents WANT to believe, which suggests it's their true preference if not their exact behavior. More revealing is the finding that even when ads cannot be skipped, viewers are less receptive to them (if "frustration" is a good measure).
@Douglas, you have a point, however another factor to consider is the tendancy of human respondents, once they are aware of the purpose or thrust of the study and agree to "cooperate", to overstate their claims and over emphasize their "opinions". In study after study, we see this overclaiming effect, which is why well designed research studies make it a point to disguise their focus or purpose as much as possible, to avoid respondent "posturing" and get a more realistic, albeit less than perfect, answer.
Could not have articulated the most important ideas better than Ed Papazian.
Hence, you're all welcome to an unsettling silence regarding the matters at hand.
Bravo, Mr. Papazian. Your thought leadership is a gift to all who have
the vision & foresight to see reality in an illuminating, constructive light.
0nwards & Upwards!
Nicholas P. Schiavone ,
Nicholas P. Schiavone, LLC
Your points are well made Ed. Once of the audience measurement systems I work on - sample rather than panel-based, self-completion - was reporting higher than expected numbers.
We did some research during recruitment that found that people that liked the media being researched were significantly more likely to agree to participate and complete the survey.
We introduced a 'propenisty weight' in order to adjust the reported results for this now-measured self-selection bias. We have also found that the propensities are shifting ground quicker than we ever expected which is now necessitating quarterly updates of the propensity weights inh addition to the regular sample weighting regimen.
@John, I'm glad to see you making such adjustments in your media surveys. As far as I am aware, nothing like that is done in our audience studies, even when independent research has demonstrated that heavy TV viewers are more likely to cooperate in TV rating panels than light viewers; the same is, no doubt, true for surveys of other media when, at the outset, the respondent---or might be panel member---is told that the research is mainly about a certain medium or, in the case of an interview, when this is obvious because the questions go immediately to the subject at hand. Add to this the "yea-saying" effect I described earlier, and you can get artificially inflated audience estimates as well as overstated time spoent estimates, etc.
Survey research to quantify volume of media usage behavior is mostly just a waste of time. It's been proven time and time again that people can't accurately report their volume of media consumption and it gets even messier when they are asked to accurately report types of activity within their consumption behavior (time-shifted, multi-platform, etc). The best we can probably ever expect to get from respondents on usage is frequency and perhaps changes in behavior but certainly not volume. For example, "Which of the following activities have you done in the past [24 hrs, week, month, etc]?" Followed up with, "Would you say you are doing [activity] more frequently, about as often or less frequently than you were [30 days ago, 6 months ago, etc]?