(For the last few weeks, I’ve been debuting a series of articles answering a question posed to me by The Ankler’s Richard Rushfield: Will The Irishman Make Any Money? It’s a great question because it gets as so many of the challenges of the business of streaming video. Read the rest here, here, here and here.)
Chatting with the esteemed Richard Rushfield a few months back—we share sensibilities on Hollywood and the (hashtag) streaming wars—he pitched me a straight forward question. Could we build a model that can answer this deceptively simple challenge:
Did The Irishman make money for Netflix?
It’s a good question because the buzz for The Irishman from critics has been so positive. From what I can tell—based on “film Twitter” reactions—this would be the greatest film ever made by man, except that with this masterpiece Martin Scorsese has elevated from mere mortal to filmmaking demigod.
It would be cool to know if Netflix made any money off it.
Which is pretty tough. I mean, we don’t even know the ratings for Netflix films…how can we determine if they are profitable? It will be hard, but to quote a famous president, we write these articles not because they are easy, but because they are hard.
So you know what? Richard and I are taking the law into our own hands. Yeah, we paint houses, with financial models and data hacks!
Later this week, in The Ankler newsletter—subscribe here for the must read newsletter—Richard will explain our purpose, reasoning and goals to start this project early. Today, I’m going start explaining how we’ll develop a “Feature Film Profitability Score”. In previous articles, I’ve pretty much built the models needed for this analysis. Now, I’m just combining them with a little special sauce.
Moreover, we’re doing all this ahead of time. We’re not judging The Irishman based on preconceived notions, but based on its actual performance. Moreover, once we build this capability, we can leverage it for future releases on many streaming platforms.
Here’s what today’s article will explain:
– The specific profitability score we’re creating.
– The four models of film release in the streaming era.
– A quick review of the traditional film model.
– Some notes on competing theatrical film models.
The “bottom line up front” is that combining my methods for valuing theatrically-released films and streaming video, we can make a model of success depending on either box office results or streaming popularity. While the last seems unknown, using some publicly available data—mainly Google Trends, potentially other third party survey data, or even Netflix datecdotes—we can make guesses on popularity.
The Goal: A “Feature Film Profitability Score”
At the end of the day, the goal is to keep this project simple. So Richard asked if I could boil this down to one (1!) number for every film—streaming or theatrical—that determines, “How profitable was this?”
Well, I failed, but I have this down to 2 numbers. Let me explain why. The obvious start is that a film can make a lot of money. This is good. Making nearly $2 billion dollars on Avengers: Endgame, Avatar, or Titanic matters. That’s a lot of money.
But you don’t just want raw totals. If it costs $1 billion to make $1.5 billion, that’s not as good of a value for investors as making a film for $200 million that makes $700 million. Same raw total, but one required less up front capital. This is a quick definition of ROI, by the way. The Joker is currently the ROI golden child of the trades. The all-time ROI club is films such as Blair Witch, Paranormal Activity, or Saw that still fill the dreams of indie horror producers everywhere.
If you wanted a quad chart of success, you could see this:
Essentially, films in the upper right are living the dream. Films in the lower right made a lot of money, but not a great return on investment. Films in the upper left made some money (they aren’t all negative), but had great ROI, meaning they were likely cheap but just not as big as some other films. And don’t be in the lower left—though most films are—which means you aren’t making money period. The majority of films in the current climate end up there. Combining these two numbers—with other metrics I’ll explain—brings us to this scorecard we’ll give The Irishman:
Beside the two promised numbers, I have four “breakeven numbers” for streaming films in particular. That’s because “breakeven” is easy for feature films (make more money than you cost), but with streaming the challenge is “what is making money”. I’ll explain those in the last section, but before we get there, we have to explain why I needed to build a new model in the first place.
The Four Models of Film Distribution in the Streaming Era
It’s no surprise that film distribution is changing. And commonly, we say, “Hey Netflix is skipping theaters.” That’s decision number one: to go to theaters or not; Netflix opts not; Amazon (formerly) and traditional studios opt in. Financial modeling wise, that’s an easy decision to calculate.
The tougher part to keep track of—and it is neglected in the media coverage—is the second window and beyond distribution plan. (I’m calling everything from home entertainment to Pay Per View to TVOD/EST to linear licensing to streaming licensing “second windows” for simplicity.) See, a new streamer like Apple is going to put its movies in theaters, but then—from what I understand—release it to Apple TV+ directly, exclusively and forever. Amazon too from now on. In other words, all these windows get condensed into this one:
The cool thing is that all the companies I think of make these two choices, meaning we have only need four models for films:
(Two quad charts in one article? Probably my favorite article of the year. Well, after this one.)
The one variable is Apple TV+. I believe they are doing streaming only, but haven’t confirmed yet. With that understanding, let’s build our models. I’ll need a model for theatrical and streaming only to evaluate the Irishman.
My “Traditional” Theatrical Model
I built and explained my first model for this site way back in May of 2018. My goal was to calculate how much money Disney made from the Lucasfilm acquisition. (It ended up being a lot). Read this article for the details, but all you need to know is that most feature films have fairly reliable waterfalls of cash flows. Here’s my estimates from last year, along with my 2019 updates:
Now, it’s been a year since I built that model and models are meant to live and evolve. I didn’t discover this Deadline feature until after I published my model, so I’ve tweaked some of the water fall numbers. Again, mainly shrinking home entertainment revenue.
That waterfall only covers costs, so now we need costs and profit sharing and what not. That is this model here:
Great! For traditional films we can use this model. (If we do this calculation for future Disney+ and HBO Max distributed films, I may tweak the model slightly—trimming some linear revenue—but that’s really only cutting a few line items.)
How My Model Compared to Other Models
Every so often, I see other numbers out there and get a pinched stressed. For example, Deadline has a great series on feature film calculations, with examples here and here. Using these, they’ve compared Avengers: Endgame to Joker, and found them equally profitable, even though Avengers: Endgame made nearly triple as much money as the Joker at the box office. I, meanwhile, would project Endgame to make much more in profit.
Seeing this big gap, I was a pinched stressed. Did I horribly miscalculate something? I don’t think so, once I dug into the numbers. First, my model is a “lifetime” profitability model and for something like Endgame—a generational hit—it will continue to make money for Disney for decades. Literally. Joker will keep making money too, but reading deadline I’m fairly certain Deadline is calculating current profit only.
Second, I lump “fees” in with the profit. That means the “fees” that Disney or Warner Bros. charge to distribute a film. While they don’t pay talent with these fees, they absolutely do get protected for the studios, and feel more like net profit than not. I’m not sure that Deadline calculates these.
Third, I still have learned some things about my model since I first built it—mainly based on how I was taught in school and saw it in action, though I’ve never specifically worked in a film finance team—which is that I might be a pinch too generous for huge box office grosses. In other words, the percentages for huge box office hits may not sustain for some Pay 1 and home entertainment grosses. When I have time, I plan to dig into some of the Deadline estimates to compare my percentages to see if I need to tweak my model.
Fourth, for something like Avengers: Endgame, which had huge amounts of talent, I probably need to make a much higher (bespoke) talent participation number. See, this Forbes article that claims that Endgame has already made $700 million in profit and paid $340 of that to talent, already this year.
Final caveat: All these models are dealing with uncertainty. Meaning that without actual finance statements, there is a margin of error. I have to make a few assumptions and naturally that means I could be wrong.