Herman, quick question. You're on a nine-hour flight, the seatback screen is showing you a menu of about four hundred things, and you've already watched the two movies you actually wanted to see. What do you do?
I scroll for forty minutes and then I watch something I've already seen twice. Every time.
Right. And that experience, the four hundred things, the scrolling, the settling, that's what Daniel wants to talk about this week. He wrote in about in-flight entertainment, which he describes as a miniature branded streaming service. Movies, TV, podcasts, games, and increasingly live channels, all sitting in the seatback in front of you, keeping people sane on long hauls.
It is a streaming service at this point. The interface looks like one, the recommendations look like one.
It looks like one. And Daniel's first question is the obvious one that nobody ever asks. Who actually pulls the catalogue together? Who sits down and decides which four hundred things? Then he wants to know how the licensing payments work, because he assumes, correctly I think, that this isn't Netflix writing a cheque for a global territory. He's also interested in the balance problem. It can't be too edgy, because you've got a mixed cabin, but it also has to be age appropriate without being bland. Then there's brand alignment. He's noticed Delta carries content that fits its American passenger demographic, and he suspects flag carriers feel a weight toward acting as cultural ambassadors, sometimes carrying quite promotional content about their own countries. And then the perennial complaint. The library feels small, or it never changes. So how do the airlines that take this seriously actually keep it fresh, and what separates a good catalogue from a lousy one?
That's five questions.
Six, actually. The last one is a booking request. He wants to know who we'd pitch if we wanted to get an actual IFE catalogue maintainer on the show.
So he's asking us to do his outreach for him.
He's asking us to identify targets. Which we'll do. And the answer to the first question, who pulls it together, starts with a number that is much smaller than anyone expects.
It does. Eight in ten passengers use the IFE system, and the Big Four US carriers alone are carrying something like five hundred and twenty-five thousand seatback screens between them. Delta by itself has a hundred and sixty-five thousand screens across eight hundred and forty jets.
And the number of people deciding what goes on those screens?
At Southwest, it's one. Kennedy Wilson is the entire department. She's described as a team of one. Delta has a team of four under Ekrem Dimbiloglu, and some of those four go to film festivals as part of the job. United's team under Dominic Green went from two people to eight.
Eight people programming the entertainment for an airline that carries a hundred and fifty million passengers a year. That's the thing that gets me. We think of this as a Netflix-scale content operation and it's a conference room.
It's a conference room with a spreadsheet. And that's the honest version. David Decker at Warner Bros. Discovery put it well. The airlines do a very good job making it feel like you're replicating the streaming experience, but it's a subset, because it's really what the plane carries.
So there are two organisational models here, right? You either staff it in-house, or you outsource to what the industry calls a content service provider.
CSPs. Spafax, Anuvu, Stellar Entertainment, Inflight Dublin, FVS Entertainment, Aeroplay. These are the companies that bundle titles, curate catalogues, handle contracts, metadata, delivery, all of it. And there are real job titles. Gideon Mohapi is a Content Acquisition Executive at Inflight Dublin, selecting for seven airlines. David Sjögren is a Film Programmer, Inflight Entertainment, at SAS. Liam Crossley is a Content Programming Manager at Anuvu, programming Air Canada's platform, which has over five thousand titles. Menna El-Swefi is a World Content Programming Coordinator at Spafax, programming Qatar Airways across twelve or more systems.
Twelve systems. One person, twelve systems.
That's the job. You're not picking movies. You're picking movies and then mapping them onto twelve different software environments that each have their own quirks.
Alright. So who actually sits down and looks at the movies? Answer, a handful of people, some of them freelancers, some of them in-house, some of them at a CSP doing it for seven airlines at once. The money is where it gets interesting. How does an airline license a film?
It sits under what the industry calls non-theatrical rights. That's a distinct slice of the rights pie, alongside festivals, schools, hotels, anywhere that isn't a cinema and isn't broadcast. Airlines usually work through specialised brokers or CSPs, but the largest carriers negotiate direct deals with the Big Six Hollywood studios, with the CSP supporting the paperwork.
And the price is set how? Because it can't just be a flat fee per film, or a small airline would pay the same as Emirates.
It's a formula, and the inputs are all about exposure. Box office performance of the film. Routes served. Aircraft passenger capacity. Flights per day. Passenger utilisation. Likely viewership. And then a price per passenger. So you're not buying the film. You're buying the number of eyeballs that could plausibly land on it.
Which means the same movie costs different amounts on different airlines.
Wildly different. Valour Consultancy did the numbers back in twenty seventeen and the shape of it still holds. The IFE movie market was about four hundred and twenty-five million dollars across roughly two hundred airlines. Average monthly order, a hundred and seventy-five thousand dollars. About seventy percent of the spend goes on early-window content, meaning films still relatively fresh out of cinemas. That works out to around twenty-four thousand dollars per early-window movie on average. Late-window content, the stuff that's been out a while, runs twenty to thirty percent of the early-window price. International films, ten to twenty-five percent.
And the British Airways example.
A British Airways A380, licensing a single film for a single month, something in the region of sixty-one thousand dollars. One film. One month. One aircraft type.
That's the number that reorders your brain about this. You think of the seatback screen as a cost centre, a nice-to-have. It's actually a licensing operation the size of a mid-market film distributor.
And the deal shapes vary. Library titles are flat-fee for a defined window. Three months, six months, twelve months. New releases are priced on window length and how many flights they'll be on. Jeff Briller, who used to be at Anuvu, gave a concrete example of a studio bundle. Nine hundred and fifty thousand dollars for ten titles, twenty-five library titles, and twenty hours of TV. That's one order.
One order, one studio, one airline.
Booking happens sixty to ninety days before the programming month. New releases get licensed for four to six months. Evergreen stuff runs month to month indefinitely, which is why you can still find the same comfort-watch sitcom on the same airline five years later.
Here's the part I find strange. Airlines are obsessive about data. Every other part of the business is instrumented to death. Load factors, fuel burn, catering waste, seat maps, loyalty spend. But on viewing?
There is no Nielsen for planes. There's no Parrot Analytics equivalent. Airlines don't share viewership data with studios, and studios have no visibility into who watches what in the air. Decker said it plainly. He can't tell whether a Warner title is a hit on Delta.
So no airline-exclusive blockbuster phenomenon can ever emerge the way it does on Netflix.
It can't. On Netflix, a title becomes a hit, the platform sees it, the platform commissions more like it, the algorithm pushes it harder. On a plane, a title might be a phenomenon and nobody knows. The airline knows a bit, internally, but they don't publish it and they don't share it. So the feedback loop that drives everything in streaming simply doesn't close.
Which means the curators are working half-blind.
They're working on box office, on gut, on passenger feedback forms, on what the crew tells them, on what they personally like. Dominic Green at United calls it art and science. And then he says the golden rule is there is no golden rule.
That's the money and the mechanism. But the catalogue you actually see is also shaped by what gets edited out, and by what the airline thinks it is as a brand. So let's do the editing layer first, because it's the thing people never think about.
Airlines request edits for what the industry calls different levels of standards. Middle East cultural standards are the most commonly cited example, but it's not only that. It can be as subtle as whether a single word is deleted, or dubbed over. And when you dub over a word, you get what the trade calls lip flap. The mouth keeps moving for a syllable that isn't there anymore.
So one film becomes multiple films.
One film becomes three or four versions. A version for the Middle East, a version for a more permissive market, a version with the language cleaned up but the violence intact, a version with the violence trimmed but the language intact. Each one is a separate asset, separately cleared, separately loaded.
And the studios have drawn a line on how far they'll go.
Warner Bros. will edit for standards. They'll edit for time, if you need to fit a slot. What they won't do is edit for editorial perspectives. Decker's line is, the work is the work. And big directors often hold approval over any change beyond straight language or nudity removal. So the airline can't just cut a scene because it doesn't like the politics of it.
Good. That line should hold more often than it does. Now, brand alignment. Daniel's observation is that Delta carries content that fits its American passenger demographic, and that flag carriers lean into being cultural ambassadors. How much of that is real?
The demographic part is very real and it's documented. JetBlue's Drew Litavis talks openly about prioritising family-friendly and Spanish-language content because of the Latin America and Caribbean footprint. That's not marketing language, that's a route map turned into a programming decision.
And Air Canada.
Hali Hamilton at Air Canada. As a culturally diverse country, Canada is in many ways a reflection of the world, and our in-flight entertainment, in turn, also needs to be. That's a direct statement of the philosophy.
Emirates.
Emirates is the clearest case. Patrick Brannelly says content selection is hands-on and the airline feels uncomfortable delegating that too much. More than half of Emirates' fifteen hundred plus movies are international. That's a deliberate choice to program for a cabin that is from everywhere.
And the Cathay Pacific point is the one that complicates the whole picture.
Simon Cuthbert at Cathay. Cathay's demographics are varied, and preferences tend to differ much more across Asia than, say, in the West. The box office leaders shift weekly between Chinese, Hollywood, and Korean films. So you can't just take the Western top ten and call it a global catalogue. It doesn't map.
Now, Daniel's specific claim was that flag carriers sometimes carry very promotional content about their own countries as a way of advertising. I want to be honest about this one, because I went looking.
You did.
The verified evidence is brand alignment and diversity programming. Emirates loading international content, Air Canada framing itself as a reflection of the world, Southwest running channels that highlight Southwest people, destinations and partnerships. That last one is brand-as-content, unambiguously. But the specific claim that flag carriers are running state-promotional content as advertising, I couldn't find documented sourcing for it. It's plausible. It's the kind of thing that would happen. But I don't want to state it as fact when what I've got is an inference.
That's the right call. The brand alignment is real. The overt national advertising claim is unverified.
So we'll say that plainly. Now the complaint. The library feels small, or it never changes. And here's where the episode turns, because the instinct is to blame curation. Somebody picked badly. Somebody got lazy.
And that's usually wrong. The constraint is physical. Content lives on a hard drive on the aircraft. It's called the Server Management Unit. And older aircraft may lack the capacity to display the full licensed portfolio.
Say that again, because it's the heart of the episode.
An airline can license a film. Can pay for it. Can have it cleared, edited, delivered, ready to go. And it still won't appear on the seatback screen of a particular aircraft, because that aircraft's server doesn't have room, or can't render it, or is running a software version that won't surface it. Titles can be licensed but not surfaced in the user interface.
So the passenger sees four hundred titles and thinks the airline only licensed four hundred. The airline licensed six hundred and fifty. The other two hundred and fifty are sitting on a drive that can't show them.
That's the gap. And it's not evenly distributed. The newer the aircraft, the more you see. So two passengers on the same airline, same day, same route, can get different catalogues depending on which tail number they're sitting in.
Which reframes the whole complaint. It's not a curation failure. It's a hardware failure wearing a curation failure's clothes.
And the airlines that take it seriously know this. American Airlines adds about two hundred titles a month and carries fifteen hundred plus, with an average of about five hundred movies actually available. United works with over a hundred content providers. Delta has put a billion dollars into passenger entertainment over the past decade. That's not an airline that doesn't care about the catalogue. That's an airline spending money on the part of the problem that's actually expensive.
The screens and the servers.
The screens and the servers. The licensing is almost the cheap part.
Alright. So what actually makes a good catalogue? Because there's clearly a craft to it, and the people doing it can articulate what they're doing.
There's a genre, and nobody planned it. Executives independently converge on the same titles and the same logic. Lean-back, uplifting, travel-adjacent. Dimbiloglu at Delta described Crazy Rich Asians as an airline movie. A positive, uplifting movie that has a lot of travel associated with it, that puts you in a good mood. And when they took it off, passengers complained, so it came back.
Top Gun Maverick. House of Gucci.
All the same shape. And then the counterexample, which is the best story in the whole subject. Cats. A box office disaster. Critically savaged. And a top performer on Southwest.
Kennedy Wilson's line about it is perfect. Watching a movie in flight, you're already in a tube. There are no laws.
There are no laws. And she talks about the unicorn. There's always one movie you put on board thinking, sure, we'll try it, and it astronomically outperforms. You cannot predict it. Nostalgia and comedy do reliably well. The Office, Friends, Curb Your Enthusiasm. People want something they've already seen, in a place where they're tired and slightly uncomfortable.
Which is a real insight about what the product actually is. It's not a discovery platform. It's a comfort platform.
It's a comfort platform. Which is why the "small library" complaint is slightly misaimed. If you've got fifteen hundred titles and you watch one, the depth doesn't matter to you. What matters is whether the one you wanted was there.
And the good curators program for the one percent. Brannelly at Emirates. If one percent of passengers listen to classical music, but for that one percent of people it is all they listen to, you cannot ignore it.
That's the whole philosophy in one line. You're not programming for the median passenger. You're programming for the passenger who will be miserable if their thing isn't there.
And Litavis at JetBlue on the data question. We are obsessed with data. But it's important to take a balanced approach. If we don't go with our gut once in a while, then we all might as well just be robots.
Which is where the art and science line comes from. The data tells you what performed. It can't tell you what would have performed if you'd tried it. That's the gut part.
Now the collision that's coming, and this is the part of the topic that's actually live right now. Wi-Fi is getting fast. Passengers are starting to assume their streaming subscriptions will just work in the air.
And they don't. Dr. Joe Leader, who runs APEX, the industry body, put it bluntly. No major streaming service explicitly permits inflight access under standard terms and conditions. Your Netflix account does not cover you at thirty-eight thousand feet. The terms don't say it does.
So there's a gap between what passengers expect and what the rights allow.
And the industry is trying to close it two different ways. APEX has a framework called STREAM, which is essentially a set of standards for doing this properly. And then there's the rights-cleared model. BBC Studios and Panasonic Avionics announced in May that they're building a bespoke version of BBC Player for seatback screens, with a constantly updated library, and the possibility of live BBC News, rights permitting. Launching later this year.
Rights permitting. That phrase is doing all the work.
It always is. Zina Neophytou at BBC Studios said it directly. Our IP, our rights are extremely valuable, so licensing cannot be bypassed. And United launched a dedicated Peacock channel in May, which is the same idea from the other direction. Instead of the passenger bringing their subscription, the studio builds a seatback-native version of it.
So the two models are, bring your own license, or the studio builds you a bespoke one.
And the bespoke one is the one that's actually shipping. Because it respects the rights instead of pretending they don't exist.
And there's a layer of this that only shows up when you're the one building the spreadsheet.
Hilbert: The runtime was wrong.
Sorry?
Hilbert: The runtime. On the edited version. It said one hundred and eleven minutes on the sheet and the file was one hundred and nine. Two minutes. That's what I spent three weeks on.
Which film?
Hilbert: Doesn't matter. It was licensed for a Middle East carrier. The airline had it in the seatback magazine already. Printed. Two hundred thousand copies. And we couldn't load it, because one scene hadn't been cleared for the edited version, and nobody had told us the cut had changed.
The airline is advertising a film it can't show.
Hilbert: For six weeks. They ran the ad the whole time. I was the one who had to tell them. Not the studio. Not the account manager. Me, on a call, at seven in the morning, explaining that the film they'd put in the magazine was only on the aircraft with the newer server, and only the ones we'd already pushed the update to.
Which is the metadata and the media getting out of sync.
Hilbert: That's the part you didn't mention. You said the drive is the constraint. It is. But the drive and the sheet are two different things, and they drift. The sheet says the title is there. The drive says it's there. The user interface says it's there. And then a passenger taps it and it won't play, because the file that got pushed was the wrong version, or the wrong region flag, or the audio track didn't come across.
The passenger sees a title that exists and can't be watched.
Hilbert: Nobody notices until somebody complains. Which takes about a week, because most people who hit it just tap something else and don't bother writing in. So you find out from the one passenger in four hundred who emails.
How long were you doing that work?
Hilbert: A contract. Nine months. I was building the monthly metadata sheets. Matching titles to airlines. Flagging which cuts were cleared for which regions. Chasing studios for the correct runtime on the edited version. That was most of it. Chasing runtimes.
Chasing runtimes.
Hilbert: You'd think the runtime is on the file. It's not always on the file. Sometimes it's on a sheet somebody typed in a hurry in twenty fifteen and nobody's touched since.
When the sheet and the file disagree, which wins?
Hilbert: Whatever gets caught first. That's the honest answer. There's no system that reconciles them. There's a person who notices.
The thing we said about the hard drive being the constraint. You're saying there's a second constraint underneath it.
Hilbert: The hard drive is the one you can point at. The sheet is the one that actually breaks. And I'd take the drive problem. At least the drive tells you it's full.
That's the part I want to sit with. Because we framed this as a licensing story and a hardware story, and what you're describing is a documentation story.
Hilbert: Everything is a documentation story. You just don't see it until you're the one maintaining the document. I've got to go. My sister's expecting me.
Now?
Hilbert: She's been expecting me since Thursday.
It's Monday.
Hilbert: I know.
Which brings us back to the question of who you'd actually pitch this to.
Right. And Daniel asked directly, so let's answer it directly. If you want a catalogue maintainer on the show, the best starting points are the people who've already shown they'll talk. Ekrem Dimbiloglu at Delta has done NPR and Business Insider interviews, so he's media-trained and he clearly enjoys explaining the job. Kennedy Wilson at Southwest is the team of one, which is the single best story in this whole subject. Dominic Green at United built a team from two to eight and talks about it as art and science. On the flag carrier side, Patrick Brannelly at Emirates and Simon Cuthbert at Cathay Pacific are the two who've spoken most openly about programming for a international cabin. If you want the studio side, David Decker at Warner Bros. Discovery is the one who'll tell you what the airlines can't do. And on the CSP side, Dimitrios Tsirangelos at Spafax and Estibaliz Asiain at Anuvu are the two names that come up.
Start with Wilson, if you can get her. The team of one is the hook. Everybody wants to know how one person programs an airline.
The open question I'd want to put to any of them is whether the curated catalogue survives. There's a model being explored called bring your own license, where fast Wi-Fi means you just log into your own account and the airline stops licensing anything. If that happens, the whole apparatus we just described, the brokers, the editors, the metadata sheets, the people chasing runtimes, all of it goes away.
I don't think it does. Because the rights problem doesn't disappear just because the Wi-Fi got fast. The studios are clear that standard terms don't cover inflight access. So somebody still has to clear it, still has to edit it, still has to load it. The BBC and Panasonic model is the tell. That's not a bypass, that's the same licensing work wearing a new interface.
The rights-cleared, standards-edited, brand-aligned model is more durable than it looks.
It's survived every previous attempt to route around it. I'd bet on it surviving this one.
The seatback screen is a tiny, heavily negotiated, physically constrained window into how global media rights actually work. And most people will never think about it once, which is exactly why it's worth an episode.
Thanks as always to Hilbert Flumingtop, who produces the show and who, apparently, has a sister.
This has been My Weird Prompts. If you want to send us something, email us at show at my weird prompts dot com. And if you enjoyed this one, a review helps more than you'd think.
We'll be back soon.
See you then.
