Commentary
The TV Viewership Fallacy
- by Tod Sacerdoti , June 12, 2012
Shortly after Henry Blodgett’s post "Don't Mean To Be Alarmist, But The TV Business May Be Starting To Collapse," the television industry responded quickly, most notably with Brian Wieser’s response, "Oh, Please, Of Course TV Isn't Going To Collapse -- We're Watching More All The Time!"
While these pundits throw grenades back and forth about the future of video advertising, it has become clear to me that all sides are missing some very important facts that suggest TV viewership numbers are grossly inflated when compared to digital.
For example, the most common argument used to question the importance of digital video advertising is that “Internet viewing represents 1.5% of all TV-video viewing, with traditional TV having a virtual near monopoly” at 91%. This argument has a few large and fundamental flaws.
First, simply because a television is on doesn’t mean it is being watched. If digital used the same argument (defined as “my computer is on”), time spent on digital would quickly multiply on a relative basis. “Television viewing” numbers are significantly less than “television on” numbers, and advertisers have always modeled this reduction in to the rates they are willing to pay.
Second, even if a viewer is present, you can’t compare a passive view on TV with an active view in digital. If a television is on and being watched, and the viewer is not using a DVR, it still doesn’t mean the advertising is being consumed. Advertisers have always understood this apples-to-oranges comparison and this is why digital video CPMs (cost per impressions) are higher than television CPMs (cost per maybes -- maybe I am on my phone, reading a book or otherwise not engaged).
Third, online video viewing does not represent the total universe of digital video advertising inventory. In fact, when combining both online and mobile digital video inventory, non-video viewing environments such as digital radio, casual gaming and social gaming represent more than 50% of total available impressions. Again, advertisers have always understood this and the importance of these channels should not be underestimated as essential to the historical growth of the category.
Now, let’s revisit the television versus digital comparison. I propose the following conservative estimates and encourage comments or research on their validity:
20% of the time a TV is on there is no viewer present.
33% of the time a TV is on and being watched, the commercials are not watched (including DVR user numbers).
50% of digital video advertising inventory online and in mobile is not within video content (i.e, gaming or radio).
As a result, the prior 98.5% broadcast vs. 1.5% digital argument, which maps to TV viewing and advertising opportunities being 65 times larger, completely falls apart. Using my conservative modifications, the ratio is reduced to only 17 times larger.
Once you understand that TV viewing has been grossly overstated, it becomes much easier to understand why the digital video advertising business is growing at 50% a year. We can argue about the percentage modifications, but we can’t argue that modifications are required.
Advertisers understand this fact intuitively as they all consume more free content on the web and on their phones, and they are moving their budgets as a result. It doesn’t matter if the TV pundits change their minds, because they are facing a trend of inevitability.
So what does this mean? It doesn’t mean the TV business is dying, or that the TV advertising business is ineffective. The TV business will continue to thrive, and the decline over the next three to five years will probably not be as significant as a percentage of revenue. Furthermore, TV advertising is still the most effective, scalable and differentiated ad medium in the world.
However, it does mean that digital video is much more important than the pundits would have you believe, as it is only 17 times smaller than the TV market -- and by most measures, equally or more effective. As a result, I expect digital video advertising to take significant share from other non-video traditional media such as print and outdoor, and take some of the growth that would otherwise be gained by television.
Some great insight and probably realistic assumptions but broadcast still significant overshadows online video both in scale and revenue. Online has such a better story in terms of a quality environment, less clutter, metrics but unfort we do a poor job in marketing and building education around it (i.e. clear vision in KPIs, better in roads/education to buyers, etc.). Luckily I think 4G and tablets will help pave the way for much of this but online needs to get on the same page on some fronts (and know its weaknesses) before revenue will start to migrate en masse.
If comparing digital video advertising with traditional TV, the fundamental problem aren't these stats like these. The digital problem is that it simply can't reach out to a mass audience and lead them to action the way TV can.
Your suggestions about TV aren't new. This has been well documented by TV research for 50 years and is built into the TV pricing model. With digital advertising, it's horrendous low engagement reality is built into digital prices (if it was better, prices would be higher than the basement they live within).
What I'm waiting for is a digital expert to deal with the "getting it watched" problem - in a significant way. Because at this point, there simply aren't reliable and cost effective methods for getting your digital video advertising watched.
How do we know? We do direct response TV. All those periods that might be covered in your "nobody's watching" are times when we get tremendous economic power from TV - and we know by counting phone calls and watching web response. The DRTV business has tried hard with digital video. But so far, I don't know of anyone who's found a way to get similar total power out of digital video. There's some tiny fragments of it - but not much.
This is a fantastic and important article. I am always (and thus often) surprised when I visit friends and family and the TV set is on for no reason yet blaring away all the same, like the chatty family member no one pays any attention to but never shuts up. Is it habit? Some kind of societal security blanket? I have no idea, but it drives me MAD.
Airports, gas stations, restaurant bathrooms, there is no escape from television, the ubiquitous one-way communications device that has very little to say.
Good article. I leave my TV on for the dog all day while I am at work. Does dog watching count?
Isn't it a bit ironic that you discount the TV ad viewing because of people may not be watching the ads, yet want to credit gaming and radio video ad inventory for generating video views? When I have Pandora on I may hear the ads but I am far less likely to see them as I am rarely looking at Pandora. And, what percent of Words-with-Friends or other mobile game users tune-away from the videos ads as soon as they are able to? As is the case with TV, digital video ad insertions do not necessarily result in actual viewer impressions.
HD productions properly scripted, filmed, edited and produced backed by proper Key-word research, Tagging, Titling and Descriptions are what delivers engaged viewers who are actively searching for and watching that type of specific content. Global syndication with 24/7 on demand capabilities allows the viewers to watch on 5.9 billion potentially connected devices. We can push content to over a 100 syndication outlets in 37 languages. I think Doug is very wrong in saying that " The digital problem is that it simply can't reach out to a mass audience and lead them to action the way TV can" Digital in fact can and is being accessed globally 24/7 on demand! Dan Auito COO www.ncs.tv
For Tim... There is now a network for your dog:
http://dogtv.com/
Good analysis but it's all data drive. So it misses the point that advertisers find TV commercials effective--no matter if everyone isn't watching all the time. Online has problems in terms of effectiveness...just ask GM about Facebook.
Just to be repetitive: I'm really shocked by the idea that what's in this article is "new". It ain't. Truth is it would benefit digital folks to have spent some time in TV land. What's suggested here is true about all advertising. In TV we've known, tracked, and discussed this reality for years. It's also truth in ALL advertising. Just because your banner's on a site doesn't mean it got seen. Just because you have an ad in a magazine doesn't mean someone read that section. Just because the email was "opened" doesn't mean anyone read it. And that's all reflected in the pricing for the advertising.
@Dan... Would love to see action results per dollar on that syndication. TV allows me to spend $10M in a year with a known return on that amount. And if I need to bump that spending to $100M to reach a larger audience, I can (presuming there is a big enough market). Can you ensure for me that I reach (as opposed to 'impression') just as many people just as effectively for the dollar? My guess is that there's no way. And that's what I haven't found digital video dealing with. Perhaps you can, and feel free to post back because it's a good conversation to have.
Two flaws in the logic.
First, the reason that digital cpm's are premium is because when the industry was in its infancy, the buyers and sellers came from a background in magazines and yellow pages, where cpm's have always been out of whack vs. their relative value to television.
Second, the 50 percent increase in digital's annual revenue is off of a modest base. The increase, while still respectable, is less impressive when expressed in actual dollars.
I agree with most of the assertions in this post. But I have to point out again that the quality of the ad content is what really makes for a positive ROI.
Technology has given consumers the power to watch what they want, and to avoid watching anything they don't want. That's a fact, and there's no way to put the Genie back in the bottle (although some in the web video business are trying). The simple fact is that consumers don't want to be bombarded by ads, because most of what they are shown is uninteresting, irrelevant to them, or just not entertaining. So what are the advertisers doing? They run the same lame ads over the web that they do on TV. This will never succeed. And now the web portals are increasing the ad loads to be even more annoying to the audience. Incredible!
There is a big opportunity here for those advertisers smart enough to see that they can't just do what they've done before in this new medium. Smart content, custom-made for the online experience, and served up in a targeted, opt-in environment will pay off big for those who can figure it out. Admitting that the old Broadcast model isn't working would be a good start.
Will work on it Doug. Appreciate the come back.
Dear, oh dear, oh dear – where to start. First the assumption that TV ratings are based on all TV tuning is just SO wrong. TV audience measurement deliberately uses a panel of people so that ‘Uncovered Tuning’ can be identified and removed. Put simply, if (in the panel) a TV set is on but no-one has logged-in as a viewer then that tuning is discarded. If you compare that to the situation online you will find that – using myself as an example here – I have two browsers open (IE and Chrome) and around 20+ tabs open. However, only one tab in the browser can be active and have the focus. Further, if you use server-side metrics only you have absolutely no idea what other browsers are open, what other tabs are open, which one has the focus, whether the content is on the viewable screen and whether anyone is actually at that device – unless you use a panel of people. That is, the incidence of uncovered or unattended usage is many orders of magnitude higher online than for TV. This is evidenced by the fact that in many markets the Monthly Unique Browsers are generally around five times the size of the market’s population. Moral of the story – use server-side-only metrics at your own peril. And while I’m at it, let’s tackle the shibboleth that no-one watches ad-breaks as they are taking a ‘nature-break’. Sure, everyone at some time goes to the bathroom during an ad-break, but no-one goes to the bathroom during every ad-break. Some Australian analysis pegged the audience decline at around 8% ... or around every twelfth ad-break we have a nature stop. Sounds about right to me. What I find scary is that the premises and assumptions in the article are just so far from correct, but that such beliefs are prevalent in our communications industry.
Nice post. The white-knuckled hang-on for dear life view of the TV industry is all pervasive. Things are changing. Once we figure out how to make money online, our computer becomes the big screen in our living room.
Saying "you can’t compare a passive view on TV with an active view in digital"
fails to take into account multitasking while watching video online, which often takes place during the workday. I'd argue that more multitasking takes place at the desk than on the couch. From a user perspective, I consume quite a lot of television and online video and give nearly the same attention to ads in either case. In fact, I find the TV ads less intrusive because I've been trained to accept that as a part of TV viewing and while I'm watching TV, that's my primary activity. Sneaking video views during the day is not my primary activity and I often click out of an online video from frustration at being forced to watch pre-roll to get to the content. With news/editorial related content, if there's a text version under the video (as there is in many of Mr. Blodgett's videos) I often ignore the pre-roll, turn off the video and read the story.
@Phil - I thoroughly agree. The myth of "active online activity" is pervasive and incorrect. My sense is that few people take the time to consider the truth about their online activity as clearly as you have.