Linear Is Just Another Bubble, Why The TV OS Is So Hard To Measure
1. Linear Is Just Another Bubble
Like many of you, I have avidly been watching The White Lotus over the past seven weeks and I am anxiously awaiting this Sunday’s finale.
The New York Times has, unsurprisingly, been giving the show massive amounts of coverage and I’ve been amazed by the number of retailers who seem to have cut some sort of deal to sell White Lotus-branded apparel.
And yet… the seventh episode of this season had a record-breaking 4.7 million viewers, up from the 4.2 million who watched Episode Six. Some of that figure comes directly from Nielsen, some from grading-our-own-homework metrics supplied by HBO, but the more impressive number— around 15 million viewers in total for each episode based on “delayed views on streaming after the episode’s debut” comes solely from HBO.
Compare that briefly to the 44 million people who watched the “Lucy Goes To The Hospital” episode of I Love Lucy back in 1953 or the 105.9 million who watched the series finale of M*A*S*H in 1983. (Adjusted for today’s population, those numbers would be 94 million for I Love Lucy and 155 million for M*A*S*H.)
Why am I calling this out?
Because it’s an easy example of just how fragmented today’s television landscape has become.
Why It Matters
Fragmentation impacts the television industry in a host of different ways.
Take measurement: we can be pretty sure that trends and patterns drawn from a series with an average of 40 million viewers per episode means something. But 40 thousand? That gives much less insight and certainty. Especially if that number is based on a panel.
Worth noting too that those numbers get even smaller with streaming—a niche show on a smaller FAST service may only get 40 viewers in a week.
Then there’s the issue of reach.
When a single ad on network prime time could reach close to half the US population, TV was a popular and easy-to-use reach vehicle. But as TV gets more fragmented, that sort of reach gets harder and harder to achieve.
Or that sort of de-duplicated reach, to be exact.
It’s easy enough to rack up big reach numbers by going heavy on certain genres, networks and services. But scratch the surface some, and you’ll see you’re hitting the same people over and over again.
That’s why finding all those people you’re missing via The Usual Suspects has been this decade’s Holy Grail. It’s made harder by the fact that many viewers just don’t interact with ad-supported media all that often. They use ad-free streaming subscriptions to watch their favorite streaming shows. Don’t do linear except for big tentpole events and maybe the local news. It’s why live sports are so popular with advertisers—the belief that many live sports viewers are people who are otherwise unreachable on TV or even on YouTube or other social video platforms.
Fragmentation is also notable in that there’s a whole lot of Culture Wars karma baked into what we watch.
CBS, for instance, has 10 shows that average over 10 million viewers an episode, based on a Live+35 metric. The top three—Watson, Tracker and Matlock, all exceed The White Lotus’s 15 million.
They just don’t have their own clothing lines at Banana Republic. At least not yet.
That said, if you’re looking for reach, they are likely even more valuable than The White Lotus.
And there are several reasons for that.
They’re ad-supported. I mean duh, but it’s notable because most of those 15 million people watching The White Lotus are watching an ad-free version of the show. In fact most (if not all) of the most popular shows on streaming have a high percentage of ad-free viewers. So if you’re looking solely at reach, all 18 million ad-supported Watson viewers on CBS are seeing your ad.
Their audience is heavily TV-centric. People who watch buzzy shows on HBO tend to be younger and better educated…and thus more likely to watch a range of media types from social video (YouTube) to mobile to web video. Meaning it’s going to be easy to hit them in other places and easy to overwhelm them by hitting them too many times. Versus broadcast prime time TV viewers whose YouTube and mobile usage is considerably lower, and so hitting them on TV is going to be the best place to reach them. Just make sure you recognize that many of them are heavy TV viewers and are seeing your ads on all the shows they watch.
It’s a desirable niche. We often hear that broadcast TV viewers are much older than the general population, as if somehow older audiences are not desirable targets for many advertisers. I mean granted, if you are a youth-oriented fashion brand it’s not an ideal audience. But for auto manufacturers, travel and tourism, consumer packaged goods and (especially) pharma, it’s a great audience.
More than that though, all audiences these days are niche. As media fragments and splinters and retreats into bubbles, different media become associated with different tribes. The good news is that advertisers who support their media—be it a podcast, a website or a TV series—get huge showers of goodwill from the audience. The bad news is that assembling enough tribes to have an audience is often an arduous process across multiple platforms where any sort of apples-to-apples comparison becomes meaningless.
Hence all the advertiser frustration.
What You Need To Do About It
If you’re an advertiser, you need to look at linear as a great way to reach a sizable portion of your user base in a brand-safe manner.
Period.
Yes they’re older and likely not as well educated or affluent. They may be concentrated in red states too, but their money is still green. In a world of audience bubbles, linear is just one of many and it’s one of the biggest bubbles out there and—added bonus—it comes with easily understood third-party metrics.
So stop saying TV is dead.
If you’re one of the broadcast networks, lean into your strengths. You have an audience that is tough to reach elsewhere. Solid metrics. Very brand safe content. Tout that. If you want to push performance metrics, well, that works too—your audience is most definitely influenced by what they see advertised on TV.
My final advice—and you’ve heard this before—is to collaborate. Remember that the enemy is GAMMA, (Google, Apple, Meta, Microsoft and Amazon) not each other.
If you’re Mike White, the creator of The White Lotus, way to monetize! I remain dazzled by the number of retailers touting White Lotus-themed gear this year. Not to mention how quickly the internet is finding all the outfits from the show. Your cultural capital game is very on point. Well done, sir. Very well done.
2. Why The TV OS Is So Hard To Measure
There have been a few reports out this past month that attempt to measure the state of the TV OS market and which operating systems are dominating it.
If only it was so easy.
In order to measure the state of the TV OS, we first need to define what we’re actually measuring.
Is it installed base, e.g. how many TVs is each OS installed on, regardless of how often they are used?
Is it how often each OS is actually used to watch TV—which one sees the most hours of viewing. Which is further complicated by the fact that many people use both dongles and the TVs native OS and not in any logical manner.
I will offer myself as a prime example of this: we have various Roku, Xumo and FIre TV devices on all the TVs in the house, and the decision as to which one to use is often based on which remote I can find first.
So there’s that and I strongly suspect I am not alone.
Why It Matters
Who controls the TV OS is no small thing. The OS is the gatekeeper of content, data and advertising. Not to mention a lucrative source of revenue.
That’s why so many companies are looking to get into the OS game and why so much attention is being paid to the battle for control of the TV OS.
But defining what we’re actually measuring isn’t easy because measuring which TV OS is in use is not easy.
We can know how many units of TVs with each OS were sold and even which households they are in. But that doesn’t tell us how they’re being used. Are they shunted away in a guest room and turned on once a year? Are they in the basement rec room where they are mostly being used as a screen to play War of Warcraft? Or are they in the living room where they’re being used to actually watch television three hours a day?
These distinctions matter and they are why the inability to accurately measure the TV OS is proving so frustrating. Media companies in particular feel this frustration as it impacts many of their decisions, from which OS to prioritize when releasing a new version of their app to which data to rely on when calculating their CPMs.
Not an easy task.
What You Need To Do About It
If you are the industry, you need to pick a metric and stick with it. Personally, I would go with the number of hours each OS is used to watch TV.
The problem, of course, being that it is a very difficult number to calculate. ACR data can tell you all sorts of things about what is being watched, but it can’t tell you which OS it’s being watched on.
Still, there must be a way to understand that metric and I will leave it to tech savvier brains than mine to figure out.
But without that basic information, the industry is going to continue to struggle to effectively take advantage of the power of the TV OS.
Which is not a good thing.
