The latest edition of Nielsen’s Three Screen report is out, and finds that US TV viewing is still growing, albeit very slowly. What’s interesting to me is that while Internet video viewing is still about 50% greater than Timeshifted (DVR) TV viewing, it’s growing much more slowly.
Here are the relevant tables of numbers from the Nielsen Wire article, but you can read the rest here.
niceee
TV in the Home includes those viewing at least one minute within the measurement period. This includes Live viewing plus any playback within 7 day; Timeshifted TV is playback primarily on a DVR but including playback on services like Start Over as well as playback from a DVD recorder.
Wait, so is the timeshifted TV only DVR playback after 7 days? Or is it all DVR playback, and those that are within 7 days are also a subset of the ‘Watching TV in the home’?
Jon K, It reads like Live+7 viewing is part of the “TV at Home” measurement, and that’s the maximum period that’s reported publicly for DVR viewing. Although I do understand that longer time periods (like Live+14?) may be measured by Nielsen, I’ve never seen any numbers for them.
According to these figures only about 1.5% of all TV viewing is done via DVR.
Here’s the math:
79.5 million timeshifters x 493 minutes per month of timeshifted TV = 39.2 billion minutes watched via time-shifting
284.5 million total TV viewers x 9207 minutes of TV per month = 2.62 trillion total minutes of TV
39.2 billion divided by 2.62 trillion = 1.5%
Kind of amazing when you think of all that’s been said and written about DVRs and how they would radically change TV and kill the business model – and certainly they’ve been impactful in Prime. But still…1.5%. Geez.
Mikey, I don’t think that’s right. To me it reads that timeshifted viewing is only viewing more than 7 days after the live program date.
I would say 75% of the shows I watch are on DVR within 7 days of the live show, but I would count as being in the first row. I don’t think you could get an estimate of DVR viewing from this information.
Mikey, I think the math is a lot simpler than that, and I think you may have made a mistake.
Avg. Viewer spends 8:13 watching Timeshifted TV out of a total of 153:27 watching TV in the home.
8 hr / 153 hr = ~5% of TV viewing is timeshifted.
That tallies pretty well with the fact that broadcast primetime is timeshifted on the order of about 20+%, but cable primetime is timeshifted less and non primetime is likely *far* less.
What Bill said. Mikey’s calculation assumes that the “timeshifters” are disjoint from the total viewers.
@ BILL: You say “What’s interesting to me is that while Internet video viewing is still about 50% greater than Timeshifted (DVR) TV viewing, it’s growing much more slowly.”. I think it’s interesting that you pull that fact out to comment on. Especially considering the increase from last quarter to this quarter is about the same for both (7% for DVR to 6% for Internet). I could look at that data and say the adoption of DVR is decelerating more rapidly than internet viewing.
I could also say that compared to last quarter tv viewing in the home has hit a plateau. It’s all a matter of reporting bias.
Hulu opened to the public in April 2008, so it is possible DVRs lost their steam after more reliable post broadcast viewing became available. It’s still increasing, but at the same time that watching tv is plateauing.
JustTuned, TV in the Home hit a plateau long ago, it has been flat or growing slowly for a very long time.
JustTunedIn says:
“I could also say that compared to last quarter tv viewing in the home has hit a plateau. It’s all a matter of reporting bias.”
I’m not sure where “reporting bias” comes in. Four data points are better than two if one is looking for a trend.
JustTunedIn, I think TV viewing has been plateauing for awhile (likely near the saturation point), but as long as the U.S. population is increasing, TV viewing should still increase.
Just threw that in as a counter point to the DVR viewing increasing more rapidly than internet viewing in the article. DVR viewing increase appears to be slowing down, but it was the main pullout of the charts that it was beating internet viewing increases, so I thought it warranted an additional comment. I may have misinterrpreted why Bill thought that was interesting. So…
@ Bill: Why did you find that interesting?
I think that DVR viewing is the greater current threat to broadcast prime-time TV network economics than Internet viewing. While it is still smaller (by about 1/3) it is growing much more rapidly.
Edit: Of course the biggest threat to broadcast TV prime-time economics is still cable network prime-time.
Ok, that’s what I thought you were saying. My point was just that the huge increase in DVR viewing was more from Q1 to Q4 08, and that the rate has dropped dramatically between Q4/08 and Q1/09. Between the last two quarters the increase was on par with internet viewing increase. We will have to wait for a few more quarters I guess, but with Hulu just out mid last year I see that as big competition to DVR viewing…. Maybe not to DVR viewing but to DVR purchasing. You can’t go back and DVR something in the past for example. I just heard about Glee today, after it aired on broadcast for example. Even if I owned a DVR I missed the chance to record it. Online I could watch it.
I still don’t watch shows at internet
Bottom line is 95% of people still watch TV the old fashioned way. Over time this will change over time. But it will likely be 10 years at least before that number drops to anywhere near 50%. Old people are living longer and longer and longer… unless National Healthcare starts euthanizing them to save money.
Traditional viewing is a habit formed over years and decades – breaking it wont be easy. Younger people will of course change sooner – unless they are too busy texting and sexting to watch TV.
Bill, check it again. The 8 hours 13 minutes of time-shifting is only within the universe of people who do any time-shifting at all. Therefore roughly 5% of viewing in homes with DVRs is time-shifted but only 1.5% of all TV viewing is time-shifted.
I think a major shift to internet viewing will occur once some company comes up with a wireless device for your computer to send internet content to your TV (or DVR thus TV) and then advertises the gizmo on TV. I know many many people that do not watch stuff on their computer screen because it is either too small or the seating isn’t comfortable as their sofa in their TV/living room. That and family/group viewing of the computer screen isn’t possible for both of those reasons.
Just this month, I corresponded with a manufacturer of DVRs about the above idea. He had posted a question to the newsgroup misc.business.marketing.moderated and I suggested the above idea. We then privately corresponded by email. He was all for the idea but afraid of lawsuits. They being sued like how p2p networks have been sued in the past. It was this sole concern that was the reason he felt they wouldn’t do it. He says all the technology is available to do it and that it won’t be that hard to produce. Cheap actually. Easy to install. Just plug the wireless transmitter into one of your computer’s USB ports and their DVR having the receiver built into their machines. It is too bad they don’t have to will to be the first to offer this, but eventually some company will. And when they do, I think it will have a massive impact on the stats this blog post presents.
Mikey says:
“The 8 hours 13 minutes of time-shifting is only within the universe of people who do any time-shifting at all.”
I’m 99-44/100% that you’re mistaken. If you look at “Figure 10″ (it’s a table) in the April 2009 Nielsen report “How DVRs are Changing the Television Landscape,” you can see DVR habits broken out by users in the “DVR universe.” It’s adults rather than P2+, but given that it’s by individuals, I think it serves for comparison.
Factoring in the decile weightings, for November 2008, one has average DVR viewing of 603 minutes per week, or about 40:14 per month, which clearly does not jibe with the 4Q08 number of 7:11 above. Instead diluting this by a rough DVR penetration factor of 30% (by *household*) yields a much closer 12 hours per month. That builds in a large fudge factor, but as a coarse sanity check, it suggests that the 8:13 is not keyed to DVD households.
Then again, I’m not getting any younger and my brain isn’t getting any less spongy, so the pointing out of gross errors is welcome.
99-44/100% “certain” that you’re mistaken. (See what I mean?)
And “DVR households,” not “DVD.” Now I’m starting to worry.