All those words and you've failed to see a simple solution to your problem. Increase the cost of the subscription service. It's that easy, even a child could come with it.
Problem there is that if you raise the cost of the baseline tier, you will inevitably lose some subscribers. It's impossible to tell the amount of subscribers you'd lose compared to the extra revenue you'd get from remaining ones at the higher price, until you actually do it.
Or, they could just add a more expensive tier. Problem with that though is that it requires more work (to add value), and you'll only get a very small fraction upgrading to the higher tier. At some point the gains in subs to the higher & higher tiers is offset by the losses in the value created for the higher tier in the first place.
What would you be willing to pay for a subscription service that had every game that's ever been made, every game that will be made and the guarantee no game will ever leave the service? I'd easily pay £70pm for that kind of deal.
Nothing, because such a service can never exist. Not one owned by a single company, anyway.
It'd have to be like a joint consortium or something with every publisher sharing the sub revenue. But then, you'll inevitably get some publishers/developers demanding they get more of that revenue because they'll try finding other metrics saying their games are bringing in more subscribers than the other publisher/developer, so on and so forth. Infighting and disagreements would probably cause that to break down.
Giving such a service to a single publisher/platform holder gives them too much power, they can just BS metrics and obfuscate data that would empower publishers providing games to the service, lowball payments for 3P games into the service, etc. Which might actually be a concern some publishers have with GamePass, I'm not 100% sure what hard metrics/numbers Microsoft shares with publishers in getting games into the service, especially Day 1 big releases, especially if inclusion into the service could impact purchasing habits of the game on other platforms simultaneously.
There's some oddities with the numbers there. Why, in this scenario, is the subscription buying out every copy that would be sold and the entire marketing budget, when the 3rd party would still have sales (both on the sub platform and outside it with marketing dollars being applied there).
That's actually a good point, I didn't account for it before. They wouldn't have to pay for all potential sales on other platforms too, you're right. But, if the game is available on those platforms Day 1, while also available in GamePass Day 1, you know there will be some loss of sales on those other platforms by gamers who gravitate towards getting it in GamePass since it would work out to be so much lower. Microsoft knows that, too, otherwise they wouldn't try netting the game for Day 1 in GamePass to begin with.
So Microsoft would still be on the hook for paying for some of the lost sales on the other platforms. Say the game sells 20% less on the other platform because it's in GamePass Day 1, and it sells 25% less on Xbox because it's in GamePass Day 1. Microsoft would still have to cover for the 20% less being sold on the other platform in some way, same way Sony has to cover lost sales of timed exclusives for not being on Xbox or Switch Day 1.
And why would a first-party game that would "normally" generate $600m in sales count against the subscription budget both for the production cost and for the lost sales? If the game never appeared in the sub, the production costs would already be removed from the $600m generated. Plus, it ignores the fact that a popular sub is going to do a lot of your advertising for you (especially in the case of first-party), drastically reducing marketing costs.
Also fair points. But to the first question, it's because the game being in the subscription will inevitably have a small impact on total sold copies. There are going to be some people who would've normally purchased it, now not purchasing it because it's in the service. So the service needs to generate enough revenue off subs in order to offset the loss in sales, even for a 1P game.
Otherwise yeah, production costs (I rolled advertising into this as well) would be a factor whether it's in the service or not, but inclusion of it in the service would impact total capacity of sold copies, and that has to be accounted for. It's kind of an invisible cost taking out from the absolute revenue the service would generate in the time frame. For advertising, the service itself being advertised can offset some of the cost to a degree, but it can't completely replace a traditional advertising route especially for a AAA game.
A lot of people are not going to bite in terms of subscribing to a service unless they know enough about the actual specific content inside of the service first, and a lot of that is still going to be done through traditional advertising, which will cost more money. So I don't think the part of advertising just pushing the service alone would cover, covers any significant amount of the total advertising budget. Maybe it does for smaller AA and indie games, but not AAA releases.
I guarantee you we're going to see a massive traditional marketing blitz for Starfield, just as a case in point. It's Bethesda's first new IP in over two decades, they're going to push this as a magnum opus across as many big channels, shows, film trailers etc. you can imagine...if they're doing it the right way