Nielsen Study Shows Vast Majority of TV Viewing is Still Done Live; DVR Viewing Beyond 7 Days is Extremely Rare

Categories: Broadcast TV

Written By

January 14th, 2013

Nielsen released its cross-platform report for Q3 of 2012 and it contained some interesting statistics about DVR vs. live viewing habits. Mainly that most TV is still viewed live and DVR viewing beyond seven days is pretty much non-existent. Check out the graphics below for more.

via Nielsen blog post:




  • Spike

    right and people dont download at all

  • DenverDean

    This is for ENTIRE day. Obviously, prime time and soaps are still DVR’d extensively based on data we see. If you think of majority of schedule (morning shows, game shows, judge shows, news, sports), all those hours will bring DVR numbers down.

  • Michelle

    What makes this data not entirely meaningful is that it’s presented in aggregate and presumably includes things like sports, reality competitions, awards shows, etc. which are probably more likely to be viewed live. We know from published DVR ratings that the +7 bump varies widely with some shows receiving a minimal increase while others might increase as much as 90-100%. It would have been more interesting and useful for that broadcast number to be broken out into sports, reality, and scripted programming and be able to compare the differences among the three. And within scripted, I also imagine there would be variances based on age group, household income, etc. I don’t doubt their findings – just feel like they’re painting a very incomplete picture.

  • Where2011

    I don’t download I stream

  • Meanie

    This graph is off. I believe the only reason Cable is viewed live more than Broadcast is b/c ESPN, NBA, NFL Network(Which are sports cause sports are 90% of the time watched live) & Award shows(VMA, BET Awards)which we see ppl watch live & reality shows like(RHOA, RHONJ, Kardashians, LHHATL, etc) cause they usual have good live numbers.

    I believe w/ 400 cable channels cable has to be DVR more cause a lot of shows come out at the same time & day.

  • Steve

    I think the real take-away here is that no matter how much advanced technology is pushed out there to consumers, most people generally don’t use anything other than basic features. I remember reading that even decades after VCR’s had been introduced, something like 90% of people had never once used them to record a show. People only used them to play purchased/rented tapes. It’s also why the “play immediately” tab on the back of the cassette was already removed on VHS tapes, so they would start automatically when inserted into the VCR. Because manufacturers couldn’t count on consumer to know how to actually push the “play” button.

    Think I also remember reading that almost all VCR users never figured out how to set the clock, which is why VCR clocks were always blinking “12:00″ in every home for years, until manufacturers figured out how to get the clocks to set the time automatically at initial installation.

    This also points out the stupidity of the fears of content owners over the years about piracy and illegal copying of content. The fear that people would be making copies of purchased movies for all their friends and family was overblown, given that the average joe couldn’t even figure out how to record a show.

    So the same dynamic is at work today with DVRs. Most (?) people have then now through their satellite/cable providers, but don’t know how to actually use it to watch content delayed and skip through the mind-numbing commercials.

  • Brennok

    They need to ask some follow-up questions in their next study. Why do they watch live? Everyone I know who watches live rents their DVR from the cable company. They watch live because they either don’t trust the DVR since they are old boxes or they don’t have enough storage to keep shows long enough to delay viewing reliably. My sister almost always watched live, because by the time she gets to the DVRed content it has been deleted for space. I think her DVR only holds 20 HD hours. FiOS only last year launched a 50 hour model which means it only uses a tiny 500GB drive.

  • Ultima

    They need to ask some follow-up questions in their next study.

    They’re not asking people questions, they’re just taking their data and presenting it.

    Also, the answer to your question should be obvious if you notice that their results were not limited to primetime.

  • Tom

    Yeah, DenverDean noted the important point right off the bat: those percentages are for the entire day. The number of people DVRing, say, The Today Show, then watching it 8 days later, is roughly zero.

    So if you assume that most of the DVRing is just of the primetime hours (3) out of what the network programs, say, 4 hours of morning shows, an hour of news during the day, and two to three hours of late night, you should be multiplying those broadcast values by 4x or 5x the numbers listed. (Multiply by even more if local news programming hours are considered in the broadcast bucket.) That pushes 1-7 day DVR viewing into a much larger portion of the pie (that more closely matches the Live+7 numbers we see), and 8+ day stockpiling into a rare but not unheard of phenomenon.

    So we are left with two points:
    1)The only number that matters is C3, the commercials watched live or within three days, which this report says nothing about. Funny that.

    2)Measuring by the entire day is misleading if not willfully deceptive. There’s a shocker.

  • David

    It’s also why the “play immediately” tab on the back of the cassette was already removed on VHS tapes, so they would start automatically when inserted into the VCR. Because manufacturers couldn’t count on consumer to know how to actually push the “play” button.

    That was a ‘write protect’ tab to prevent accidentally recording over your movie. Some VCRs may have been designed to start playback automatically on those tapes, but that was not the purpose of the tab.

  • Ray

    Is OnDemand viewing considered part of DVR or is it not included at all?

    My viewing patterns are almost the opposite. I was out of town last night and hadn’t recorded The Mentalist (which I DVR because it’s not available OnDemand)so I watched it live. I know it’s the only time I’ve watched a live episode of scripted TV this year and probably in the last three.

    OnDemand is great if you can hold off for a week and avoid spoilers. ABC and NBC, which disable fast-forward, reduce the OnDemand ads by 10-14 minutes after a week, and at least one (TNT) enables fast-forward again. FOX is the network that tends to keep its shows a full 30 or 60 minutes for the entire OnDemand run. CBS is the most viewer-friendly, with very few ads from the start and fast-forward enabled, though they and FOX usually only keep the last four episodes, whereas with NBC and ABC it’s five or more (ABC changed their policy recently – they used to only show the last three episodes). CW just started OnDemand with Comcast (but not Time Warner) and I’m pretty sure I was able to fast-forward theirs. The cable offerings OnDemand are also pretty viewer-friendly.

    I just can’t imagine being tethered to the couch between 8 and 11 every night.

  • Dan S

    I DVR most scripted programming. I watch award shows & the news live which makes little sense to record. I did watch The Simpsons live last night only because I was watching part of the Golden Globes pror to that & had no programs to catch up on. I then caught OUAT & REV already in progress & zapped the commercials.

  • Anna Bones Clarkwood

    How often does Nielsen change who has the boxes? Do they get them for life as long as they don’t move (or even if they move), change the family dynamic (kid goes to college/new baby who gets to be 2+/married/divorced) etc?

    I wonder if a random sample of say 70% of their box data comes really close to 100% (which I assume is what they normally use) of their box data. Do they use 100% of their box data or do they use less than that so the randomness of it all is more accurate?

    To me, while they try to have every demographic represented well, their sample size, while larger than presidential election polls (I’m told) is still so small (Presidential is really only 2 candidates/ 1 main question), that for example, “Latino male aged 18 with an income of $22K living in MPLS/SP, MN all by himself”, is probably only represented by 1 or 2 people. How can that sort of sampling be worth much of anything?

    I understand that the advertisers use/trust Nielsen, so for now, that’s the status quo. We also have Government Organizations that are the sole source of information, like the FDA. And, while I have to trust the FDA, I think Nielsen should up their Boxes to 250K or more while also giving out the average of 5 random 70% sample data. That’s still a far cry from 315mil+, of which, 98% own at least 1 tv. Yes 315mil+ is total ppl not households which is what Nielsen does. I wonder if they have multiple Boxes in some HH? They def should for a proportionate # of those that have tvs in other rooms (kitchen, bedroom, second family room, garage etc).

    Bottom line, just cause Nielsen is the only real source & thus we (advertisers, viewers, networks) have to live with doesn’t mean that their data is actually accurate or even comparatively among all tv shows. Nielsen is just a better/more scientific twitter/facebook circle of friends than our own. However, it’s not as good as it should be. And thus, their circle of friends does not have to equally represent the true US Nation as a whole.

    So, yes, Nielsen is the best we have going now, but that doesn’t mean that a decent number of real questions about the validity of it’s data is unfounded.

  • Freddy Arrow

    1. I thought I watched a lot of TV compared to the average person, but I’m no where near 4.5 hours a day. I suppose there’s been some rare week when I watched over 30 hours, but its hasn’t been very often.

    2. About half of my viewing is of content that has been on my DVR for over 7 days. I guess I’m in that extremely rare segment of the population. Of course, only about 1% of my viewing is live, so that shows how different I am from the average.

  • chezmoi

    still not made clear the age of the persons (ie eyeballs) watching the tv or dvr

© 2015 Tribune Digital Ventures