Well, last week we had pretty big news: the Warner Bros (formerly Warner Media) and Discovery merger officially closed. I had covered this topic in-depth when the deal cleared the final regulatory hurdles in February, if you want my thoughts.
But what about the political machinations of last week, like the departure of some key Warner Bros execs? (Jason Kilar and Ann Sarnoff in particular…) That MUST be the story of the week, right? Wrong. But I’ll explain that in next week’s column…
Instead, I think it’s time to update my proprietary estimates for the number of subscribers in the U.S. for each streamer. Because now that Warner Bros merged with Discovery, the combined Warner Bros-Discovery has a much more impressive subscriber total.
Here you go:
As a reminder, here is what these estimates are:
What – Number of paid subscribers
Where – In the United States (not with Canada!)
When – As of the end of 2021 (the end of Q4, or December 31 2021, roughly)
Who – From each major stream (over 5 million subscribers)
What – These are my estimates, based on financial reports, leaks to journalists, other analytics firms estimates and my models.
Actually, I should also give you the “why” for this chart.
First, I focus on “U.S. subscriber” estimates instead of U.S. and Canada, since most folks don’t think of TV as a U.S./Canadian market, and they really do have much different regulatory environments in television. (For movies, most box office is “domestic” meaning U.S. and Canada.)
But why this market in particular? First, as an American, the world revolves around us. (Kidding.) Really, the U.S. is the most mature market where every streamer has launched. And comparing global subscriber totals between services is frankly baffling to me. That’s about the least apples-to-apples thing you can do. (Plus everyone from Variety to Axios already does that. So where is the value add there for me to simply repeat those numbers?)
Second, my focus is on “paid” subscribers, since it is easy to game “all” subscribers. For example, Apple and Amazon almost give their service away. And from a strategic value as a company, that may make sense. As a metric of success, though, it isn’t a sign of good strategy. Good strategy is offering a product customers want to pay for.
If you want to see the above chart in table form, here’s that:
The colors relate to my certainty of each number:
First, here’s a great quote from Children of Dune for all the newish Dune fans out there:
If you believe certain words, you believe their hidden arguments. When you believe something is right or wrong, true or false, you believe the assumptions in the words which express the arguments. Such assumptions are often full of holes, but remain most precious to the convinced.
-The Open-Ended Proff from The Panoplia Prophetica
Since only two streamers release their actual paid, U.S. subscriber estimates to the public, these numbers are my estimates. Emphasize estimates. Which means, yes, there are a few hidden arguments (assumptions) I want to explain.
(Do you just wish each streamer provided their total subscribers quarterly? Me too. Personally, I think the trade unions—the DGA, WGA, SAG-AFTRA, and IATSE—are best positioned to insist the major streamers provide these numbers. Them or the SEC.)
Now, as I explain these assumptions, you may disagree. Go ahead! To quote the dude, these are just like my opinion, man. In fact, I’ve done my job even if you disagree, because I probably gave you a number to anchor your judgement, so you can adjust it either go up or down on. That’s useful!
Here are my quick assumptions by company, starting with the easiest estimates to the hardest.
This one is easy, since Netflix provides U.S. and Canada numbers in their earnings reports. (UCAN is their acronym.) They used to only provide U.S. only numbers, then they provided four quarters of U.S. and UCAN data. The U.S. was about 90.3% of the subscriber total. If you just went via population, Canada is about 10% of the combined U.S./Canada population, so this number makes sense from two different angles.
Disney/Disney+ & Hulu & ESPN+
First, Hulu and ESPN+ are U.S. only, so those estimates are straight copied from Disney’s earning reports. Disney started provided Disney+ U.S./Canada subscriber totals in their last earnings report, so I use the same Netflix multiple (90.3%) for Disney+ in the U.S.
For Disney’s “total” subscribers, though, I want to take out “bundle” customers, so I make an estimate of the number of bundles they’ve sold (currently 15.3, based on growth of Hulu and ESPN+ since the bundle was launched) and subtract that twice from the total subs per service. (This is why the Disney Total isn’t simply the summation of all three services.)
Lionsgate provides these U.S. numbers directly. Thank you Starz!
One assumption I do make, though, is that I value premium linear subscribers as much as streaming subscribers, since they could easily transition to streaming. I do think there is a difference between say the folks who subscribe for the cable bundle as a whole and those who pay for premium services as an add-on. So for HBO, Showtime and Starz, I count these subscribers with their streaming counterparts.
Previous to 2021 Q4, Peacock’s numbers were mostly my guesstimates based on vague leaks. BUT!!! The great news is that in the last earnings call, Peacock mentioned the number of paid subscribers. Since Comcast is often fairly opaque, I doubt they do this again—and I’ll be back to estimating based on presumed growth after this—but for this quarter we have a reliable number
They provide a total of all their U.S. subscribers, though it isn’t broken out by streaming service. Part of me thinks they have an outside shot to take a seat at the streaming table, so I include their total number, though technically no single AMC-owned streamer likely has more than 5 million U.S. subscribers
Warner Bros/Discovery/HBO Max & Discovery+
The good news? AT&T provided HBO Max U.S.-only subscriber totals. Like Starz, this includes both premium linear subs and streaming-only subscribers. I think this is fair, but you may disagree. Good luck untangling who counts as an HBO Max sub versus and HBO sub.
The bad news? Discovery has only provided Discovery+ global numbers. Grrr. In this case, I estimate that after the initial quarter, half their growth is global. That’s an assumption, but it’s the best I can do. Next quarter, it will be fascinating to see how these number change via Discovery’s reporting.
Paramount/Paramount+ & Showtime
Man, even keeping the name of this company and their streaming services is a chore. So is trying to figure out their U.S. subscribers.
Between CBS All Access and Paramount+, we’ve had a host of leaks and different calculations/definitions in their investor disclosures, switching between U.S. and global numbers. So Paramount (formerly ViacomCBS) has a bunch of data out there, but it’s hard to make sense of it all.
Paramount last said they have 56 million global customers, and I assume 50% of those are in the U.S., between Showtime and Paramount+. I used a few historical ratios to estimate the current total (15.1 million Paramount+), but this number could be much higher or lower. (Meaning, if I change these estimates significantly in the future, as I refine estimates and get more data, don’t be surprised.)
And now we’re into the guessing game part of the estimates. Amazon rarely provides data on streaming. Previously, they’ve said they have 200 million global prime members. They’ve also leaked that 175 million folks “watched” a video via Amazon in a calendar year.
In this case, using the number of global prime member (I estimate it at 250 million members globally) and the number of global users as of January 2020 (175 million), I assume 2/3rds of global subscribers are in the U.S. (116 million) and 30% use the service monthly. That gave me 35 million subscribers in Q2 of 2021. Then, I assumed 5% growth from Q2 to Q4. To double check this, I compared Prime Video usage to other services, using Nielsen’s monthly The Gauge.
(Also, post-merger, I realize Epix is now part of Amazon. We’ll see if they disclose any numbers in the future.)
Similar to Amazon, Apple doesn’t provide global or U.S. numbers and provides folks with three free months of streaming. (Initially that got extended to fifteen months.) Since Apple doesn’t provide numbers, I’ve used a few leaks to triangulate this figure. In the middle of last year, there were leaks that Apple had about 40 million global subscribers, of whom half are “paid”. Another leak (when Apple was negotiating with IATSE) revealed that Apple has about 20 million U.S. subscribers, about what I estimated last year. So assuming about half of those are paid, and using my previous estimates, I peg that Apple has just under ten million paid subscribers currently.
(This number has the biggest chance to move around given new data.)
A few weeks back, I spent a good chunk of time updating my subscriber estimates. Previously, I’d updated the numbers in Q3 of 2020 and Q2 of 2021. My goal is to keep these updated quarterly going forward. (Wish me luck!) I still have a few more sub-projects related to U.S. subscribers, including updating my estimates for the quarters I haven’t covered, making a timeline of subscriber totals over time, and comparing some of my initial estimates to updated data points.
Longer term, I’m thinking about adding price of streamers, ARPU, usage and other metrics I bundle under “usage”. Always be improving!
Finally, you may be thinking. “Hey, it’s February. These are late!” Well, not really. Most companies provide updated subscriber data when the release their earnings reports, which start about three weeks after a quarter closes and goes for up to eight weeks after. So I could have put these numbers out in March—and I did over at The Ankler—so I’m only about four weeks late. Around June, I’ll provide an update on these numbers.