If you ever found yourself in the armed forces, you may be familiar with the phenomenon of “hurry up and wait”. It worked like this. A dignitary is coming in. Say a three star general. So your two star general plans a division parade. Which means the Command Sergeant Major sets it up. Scheduled to start at 1000 (ten hundred in parlance), the Division Sergeant Major tells his Brigades he wants them in position by 0930, so of course that means standing in place at 0915.
Woe to the unit that is late, by the way, which is why everyone is “hurrying”.
Here’s where the fun starts. The brigades don’t want to be late for the 0915 “time hack”, so they have their troops show up in formation at 0830, to be ready 30 minutes early too. Then the battalions say, well, “Let’s be ready 30 minutes before that.”
And so on. With companies, platoons and even squads.
Eventually, you have thousands of soldiers leaving their houses at 0500 in the morning—and skipping PT— so that they can stand in formation for 3 hours doing nothing. Hurry up and wait.
That’s my feeling for Netflix’s latest earnings report. If I can speak for my fellow Netflix hawks—be they bulls or bears—we had this date circled since July. Will Netflix miss subscribers targets again? Could the stock tank? Or will Netflix crush estimates? And what will Reed Hastings and Ted Sarandos have to say about the “streaming wars”, which start next month?
No kidding, the (hashtag) streaming wars made it in the shareholder letter.
After all the waiting…we didn’t learn much. Netflix hit almost all their targets on the dot and Hastings/Sarandos didn’t reveal any ground breaking news.
Instead, we hurried up, but we’re still waiting for the streaming wars. Sigh. Still, we learned a few things in the latest earnings report. Let’s start with strategy, touch on the specifics and end with the subscribers/financials. Oh, and meta thoughts to conclude.
Strategy: Focused, but Narrow?
When it comes to strategy, Netflix probably has the clearest plan and simplest business model of all the streamers. They sell subscriptions and they plan to dominate all scripted and reality viewing. If you want to know their strategic advantage, it’s that. They know what they are, and they’re extremely focused on that.
And they still have a big goals: they want to take over TV all over the globe. If you want Netflix’s global upside for share price, it’s all this. Honestly, “Netflix bears” like myself probably neglect how much additional revenue can come from this global conquest.
But it isn’t cheap, of course. Netflix is spending $10 billion in amortized and $15+ billion in actual cash each year (from 10K) to get those new subscribers. Worse, the most valuable subscribers are in the United States, then Europe, then Japan and then descending down the GDP per capita income ladder. And those most valuable markets are nearing Netflix saturation. (Meanwhile some, like China, remain off limits.)
The most interesting observation, for me, was about the content strategy. Here’s the introduction to the content section of the shareholder letter:
This content strategy manages to be both broad and limited, at the same time. Broad because Netflix is competing on film and TV and scripted and unscripted for kids and adults and everyone.
And it’s limited because that’s all they are competing on? If “live TV” isn’t an option, that means events, sports and news just aren’t in the cards.
In other words, for all the growth in content spend, many TV consumers will want a TV service that provides news and sports. If Netflix can’t fulfill someone’s entire TV demand, that means customers will have to watch other options too. If you value Netflix’s stock price as “becoming TV”, well you’re seeing a company that knows it won’t be that. Sure, it has a chance (that is becoming smaller every week) of “become all scripted and reality TV”, but that will be a pricey fight.
My view is that as long as customers need to have other TV services, Netflix is increasingly likely to be churned out of someone’s regular diet.
Nor does Netflix have the capital to branch out into a large new business like either buying DAZN or Roku or even a Lionsgate. They could merge with any of those, but any new venture that has huge upfront costs will stress their cash flow even further. Apple, Amazon and Disney don’t have those hindrances. That means Netflix needs to be perfect on everything decision from here on out to maximize that share price. That’s tough.
So let’s get into those specifics.
Content: Some Selected Datecdotes Tell Us Some Things
Let’s start with the headlines you did not see yesterday. And the “null results” because I want to place in your easily available forebrain the idea that most shows aren’t doing well on Netflix. Or they are and it behooves Netflix to keep them secret. (I guess.)
Of 162 original launches in Q3, 93% did under 29 million viewers. They were bombs.
Fine that is a bit provocative. Technically, I don’t know how many series really had less than 29 million viewers. Just because the lowest Netflix datecdote was 29 million doesn’t mean some other movie secretly did more than that, and Netflix just didn’t say. (Which, why hide that?) How’s this:
We don’t know how 93% of Netflix originals performed last year.
It’s crazy to consider that’s how many launches Netflix had in Q3, but according to my data—which means sorting this long, long list from All Your Screens Rick Ellis—that’s accurate. And given that when we do get inadvertent leaks—Steven Soderbergh said in an interview High Flying Bird had 8 million views—they are much lower, I’d bet that most other content releases were much lower than the leaks.
On to the datecdotes (explanation here) we did receive. These are both higher and lower than it seems. On the high side, I counted 12 datecdotes (authorized or leaked) since Q2. No seriously, four movies and 8 TV series. The challenge is that four of these datecdotes apply to two series (Netflix released 7 and 28 day numbers for both Stranger Things and La Casa de Papel.) And three of the TV series releases didn’t actually have numbers just “biggest TV series in the country” (Sintonia, The Naked Director and Sacred Heart.)
Since I’ve done feature films the last two times, I’m going to try to draw out some conclusions from the TV side this edition. Here’s my summary table of the data:
And now some implications.
– This is the problem with datecdotes. With Nielsen, you get the same data all the time. In the same formats. With Netflix, even with more data leaks, you can see how they started dropping in seven day viewing (ST3 and CdP) and lifetime customers (for Orange is the New Black) and now “biggest in country X”. It makes my job tougher and is not apples-to-apples. (For a data company, you’d think they’d know how important apples-to-apples comparisons are!)
– The drop off for TV seems even steeper than film. 40 million folks tuned in for Stranger Things season 3; after 3 more weeks that number “only” rose to 64 million. So 56% of viewing was that first weekend. But La Casa de Papel was even worse: 76% came in the first week. Compared to the two data points we have for films, Bird Box had 56% of viewing in the first week and Murder Mystery had 62%. This could be noise, but I have a feeling the drop off is steeper for TV than films.
– What about the logarithmic distribution of returns? It likely still applies, but finding the denominator for TV has been tougher than film. Moreover, with only 9 data points, we still lag a bit behind film, which had even sharper distributions. I also think that the “70% of one episode” metric is much noisier than series completions, which would show even more exponential returns.
Those are the numbers, how does that mean Netflix is doing?
Generally, I’d say yes they had a better quarter, but still have room for growth in quality content. Here is a table I made last quarter to show this:
You want a visual? Fine, I figured out a way to do that:
I like this look because even though the total is higher, it shows that the cluster was a little tighter in Q3. In other words, Netflix didn’t have a breakout hit film. (In Q4, Netflix is banking on El Camino (the Breaking Bad film) and The Irishman both being “Bird Box”-level hits. And The Witcher doing Stranger Things-type numbers. I’m not convinced The Crown, buzzy though it may be, is a true popular hit. Curiously, we haven’t gotten a datecdote after the opening weekend of El Camino which mean it isn’t doing well, or they just held on to it.
As for how popular these things are in the United States, well here’s Google Trends:
In other words, there are hits, and then there is Stranger Things.
Another crazy data point: in Q3, across all platforms in the US, 800 “things” were launched. Holy crap. That is “too much TV”. We are definitely in a content bubble.
The last elephant in this room to tackle is the international flair of this week’s earnings report. From Sacred Heart to Casa de Papel, the point was growth is coming overseas and that’s what Netflix is delivering. This topic is so unique and worth explaining that…it will be its own article. Sorry.
Distribution: More and More Cable Distribution
Here’s the upside for Netflix: They’ve kept expanding into MVPD (cable or satellite) distribution. Here’s the key quote from the shareholder letter:
Honestly, this will be the hardest part—in my opinion—for Disney+ and Apple TV+ to catch up to.
Let’s get this out of the way: subscribers aren’t financials by themselves. They aren’t money. They aren’t free cash flow.
They are, however, a tremendous indicator of potential free cash flow.
I bring this up because there do seem to be analysts who begin and end with subscribers. Check the sub numbers, if they went up, click buy on the price target.
My counters are two fold. First, as Matt Levine has parodied in his newsletter, it is almost as if some businesses now have the plan to add subscribers, make a business later. He quoted Kevin Roose of the New York Times as both were lampooning Movie Pass and it’s money losing ways.
He’s not wrong. So before I talk about the subscriber numbers, we have to remember they represent real people buying a product (or forgetting they are paying for a product), not just a number on a spreadsheet. Also, high subs now aren’t high subs forever. AOL much?
The good news is Netflix “hit” their sub numbers (or just missed). In the US they added 500K subscribers and are forecasting 61.2 million for next quarter. For those keeping track at home, this would be about 1.2 million above my 60 million forecast last January. Though a few months after that forecast, I sort of wished I had kicked it up to the 62 million range, and 61 will be right between them. In all, that would adjust this table to…
…these updated numbers. This actually would lower my Bass Diffusion estimate. In my original model, I went high and assumed Netflix would end Q4 in the United States much closer to 62 million. Now even Netflix thinks 61 million is more likely. Plus, Domestic DVD subs came down more than I had forecast last quarter, which shrank the overall subscribers line.
Either way, it’s looking less likely that Netflix smashes through the 70 million US subscriber threshold.
More disturbing to me, though, is the fact that I had to add another color to this already rainbow table of numbers. Next quarter, Netflix is going to start combining the US and Canada numbers into “UCAN”, which means this table immediately becomes out of date. (Or I’ll have to back into a US numbers.)
This drives me bonkers as I ranted about on Twitter. The simple solution is that when they make this change, Netflix could just provide ten years of Canada subscriber data. That’s not that hard to do, is it? Just pull it from your internal files. Then I can add them together for a new look.
But they won’t, because less data for investors is better for Netflix. Which is why I want a strong SEC to make a level investing playing field for all Americans. But that’s a political point for another time.
Finally, the other revenue and importantly free cash flow numbers stayed on their current paces. I still think that “free cash flow” represents the biggest risk to Netflix’s share price if they can’t stay on the growth track they are on. So each earnings report, that’s the two numbers I scroll to (current losses and next year’s projections).