Xbox's Restructuring Isn't an Asha Problem. It's a Game Pass Math Problem.
Microsoft can't seem to stay out of the news, and this week it earned the headlines the hard way.
A leaked internal memo, a fresh round of studio closures, Ninja Theory reportedly shutting down, and midsize studios told to find a buyer or get wound down. On this week's Player Driven Live, Greg Posner and Colan Neese spent most of the hour on a single question that matters to anyone working in live ops, monetization, or platform strategy: was this restructuring a leadership failure, or the inevitable bill coming due on a business model that never penciled out?
The short version of their answer: stop blaming the person who walked in last. The decisions that produced this week were made years ago.
Game Pass borrowed its playbook from Amazon Prime. That was the mistake.
The most useful framing in the episode wasn't about Xbox at all. It was about where the Game Pass strategy actually came from.
Neese, who spent time at Amazon, laid out the lineage directly. Amazon spent a decade unprofitable on paper while it poured money into Prime, and it kept Wall Street patient with one promise: subscribers don't churn, so lifetime value stretches into infinity. The stock held. Then AWS revealed its real margins and the multiple exploded. The Prime-LTV story worked because there was a cash engine underneath it.
Microsoft, the argument goes, ran the same play with Game Pass without the engine. Grow subscribers, tell the street the lifetime value justifies the spend, and assume those players stay forever. Every entertainment company chasing Netflix's subscriber multiple made a version of the same bet. The problem is that you can't build a balance sheet on engagement numbers and the presumption of infinite retention. When subscriber growth stalled, the whole structure folded in on itself.
The line worth forwarding to your team: this was never really an Xbox-versus-Sony story. It was a subscription-economics story, and gaming was the last category to learn the lesson the streaming industry already paid for.
The studios on the chopping block were set up to fail by the model, not the games
Posner ran through the list of titles Microsoft has flagged as underperforming: Avowed, South of Midnight, Hellblade II, Forza Motorsport, Outer Worlds 2, Ninja Gaiden 4, and more. His observation is the one that should sting for anyone in live ops: most of these are niche, buy-it-and-finish-it games being judged against an engagement metric they were never built to hit.
That's the trap. When success is defined by engagement, and engagement is what feeds the LTV math, a beautifully made single-player game with a beginning and an end looks like a failure the moment players finish it and leave. Neese mentioned a former Forza leader who fought this internally years ago: Horizon used to ship two DLCs a year, and the mandate became "think like a live-service game" with live ops, live content, and a recurring cadence. Not every game can absorb that. Not every game should.
There's also a marketing failure layered on top. Outer Worlds 2 got a half-hour Summer Game Fest slot that didn't move the needle, while Black Ops 7 — the title that actually needed the airtime — didn't get it. Posner's read: bad games get sold all the time, and marketed games sell copies. Several of these launches got neither the model nor the marketing they needed.
The $300 million Call of Duty question
The sharpest disagreement in the episode was over Call of Duty, and it's where the economics get concrete.
Posner's case: Call of Duty was historically the number-one selling game of the year, every year, good entry or bad, because Microsoft charged full price and "call of duty dads" bought it on habit. Plenty of those buyers play ninety minutes and shelve it, and that's fine, because money is money. The problem is what happened when the game went free on Game Pass everywhere except PlayStation. By his read, Black Ops 7 was a monster hit on player numbers and still reportedly lost around $300 million, because you can't recover a full development budget on engagement when you've given the unit away. You can't make up cheap money out of thin air, especially when the parent company is hunting for cash to build AI data centers.
His prescription is the one he's written about repeatedly: windowing. Sell the premium game at a premium price at launch, then window it down to the subscription tier a year later, the way Sony, Nintendo, and Disney all do across games and film. Day-one-on-subscription trains your customer not to buy, devalues the IP, and teaches players to try a game for five minutes and move on.
Neese pushed back, and the tension is real: you can't credibly sell exclusives and offer a free ad-supported tier on the same platform with the same message. Trying to serve a premium audience and a free audience at once just deepens the confusion Xbox already has. Posner's answer was windowing as the bridge between the two tiers rather than running them in parallel, but neither host pretended it was a solved problem. That's the part worth bringing into your own strategy conversations.
Where the money goes next: ads, targeting, and a long tail of small studios
The forward-looking stretch of the conversation is the one operators should sit with. Neese's argument: once you discount access to content toward zero, you monetize attention through ads, and the companies that get serious about an ad engine turn that attention into a flywheel. Meta, Google, and Amazon already run ad businesses north of $100 billion a year; Reddit, Snap, and TikTok monetize free access the same way.
The non-obvious part is that the same engine that targets ads also targets content. TikTok's content recommendations are as good as its ad targeting because it's the same machinery. Applied to games, that means discovery stops depending on throwing money at Twitch streamers and praying. A small studio can find the core thousand players it needs to sustain a business, because the engine can put the right game in front of the right person. The upside Neese sees over the next five to ten years isn't bigger blockbusters. It's a viable long tail of small, sustainable studios that never had a discovery path before.
Quick hits from the Summer Game Fest top 25
Neese also previewed his Summer Game Fest mindshare ranking, combining week one and week two scores across the four events that actually broke through: State of Play, Summer Game Fest, the Xbox Showcase, and the Nintendo Direct. A few of the unreleased-game standouts:
- God of War (No. 1) ran away with it, roughly 3x the mindshare of the number-two title and barely a week-over-week drop. The methodology tracks consumption, not sentiment, and the argument for ignoring sentiment is blunt: negative-sentiment games tend to crater 70 to 80 percent week over week because the internet only stops for a car crash once.
- Fable (No. 2) surprised both hosts given everything else happening at Microsoft.
- Zelda: Ocarina of Time remake (No. 3), Resident Evil Requiem (No. 4), and Call of Duty: Modern Warfare 4 (No. 5) rounded out the top five. The hosts flagged a potentially brutal February if Fable and the next Resident Evil land in the same window.
The line that closes the episode
Both hosts kept circling back to the same uncomfortable point. The executives who made the calls — Game Pass strategy, the Activision price, day-one-on-subscription — have largely moved on, and they did fine. The people now holding the bag are the artists, writers, designers, and engineers who were executing a vision someone else set. That's the part of every restructuring that doesn't show up in the mindshare data, and it's the part this audience lives closest to.
Get the Tuesday briefing.
Three things every Tuesday: the week's best essay, the next live drop, and one thing we're still arguing about internally.
Ready for more? Join the Discord community or see the Academy.
Open the latest Intel drop.
Source-toggle Newzoo's top-20 charts, see the all-four club, run the money-versus-attention map. Player Driven Intel is the data side of the show.
OPEN THE INTEL DROP