EU Plans to Let Spotify Direct Users to Other Payment Options on iPhone

[Abazigal]

You accused them of doing shady stuff with Spotify - when in fact they did what Apple should be doing aka let them not have to pay that fee. And I didn’t say you defended Google once…

Nothing misinterpreted - you tried to defend Apple using Google, incorrectly, because you failed to understand the facts of both cases.

20 things we learned from the Epic v. Google trial.

Okay, technically it’s a little longer than a fortnight.

www.theverge.com

What Google doesn’t say, and I weirdly didn’t hear Epic suggest either: if some companies get sweetheart deals, everyone else effectively has to pay more to compete. Can you imagine trying to compete with Spotify with your new music app while paying Google 11 percent, at the same time Spotify is paying zero?

The reason why I feel what Google did with Spotify was shady (waiving their 30% app store cut) was shady is because Spotify is not the only music streaming service in the App Store.

Apple is consistent with all the various streaming services. They all pay 30% (or 15% after the first year). You can argue that Apple unfairly favours its own Apple Music service, but at least it's consistent with everybody else.

For Google, how can anyone argue that it's fair to have Spotify pay 0% while every other service, from Tidal to Deezer, would still need to pay their share? This effectively gives Spotify an advantage over the competition because they have lower costs, which in turn translates to more operating revenue to run their business over everything else.

So it's funny that Spotify complains about Apple Music having an unfair advantage over it, yet they have no qualms about receiving the same preferential treatment from Google when the opportunity presented itself.

Hypocrites.

 

[webkit]

Calculate it. Spotify has 500 million users. Assume 100 million users is on iOS. App is 150 MB. Spotify pushes 4 updates a month.

That's 60 Petabytes every month from just Spotify. Based on standard S3 pricing, that's $3 million per month. Spotify no longer has inapp purchases so Apple loses $3 million/month from one app.

And that's just *bandwidth* cost. Didn't even get into server costs, TestFlight costs, CDN costs, app thinning costs, app review costs, etc...

Your $ calculations seem way too high. How much do you figure it costs Apple to run/maintain the App Store per year?

 

[jlc1978]

As a matter of fact, I’ve bought many macOS app from non-Apple stores at prices lower than the Mac App Store’s.

Yea, it seems there is. lot of variability in ricing. I've bought a number on sale as part of bundles, while others are the same price.

It will be interesting to see how pricing is impacted. iOS apps already tend to be less than MacOS prices, so I suspect developers will try to keep the extra 30% in many cases since their margins are small already and the added costs of hosting, payment, etc. will eat into them; as well as potentially fewer sales outside the App Store.

01
Apple Charges a Discriminatory Tax:
Apple requires that certain apps pay a 30% fee for use of their in-app purchase system (IAP). However, the reality is that the rules are not applied evenly across the board.

Spotify's choice of words is telling. They use "tax" because no one likes a tax; had the used commission of markup, which s more illustrative of what Apple does, it would not elicit the same response.

Interestingly, Spotify charges rights holders a 30% tax as well... In addition, they keep 100% of the revenue if you don't hit a minimum threshold. Seems they are not applying their rules evenly, either.

Does Apple Music pay it? No.

Hard to say how Apple Music is accounted for on Apple's books. They very well may pay a 30% transfer payment to the mothership as part of their P&L.

Does Uber pay it? No. Deliveroo? No. Apple does not compete with Uber and Deliveroo. But in music streaming, Apple gives the advantage to their own services.

Apple has been clear that providers of tangible goods and services do not pay a fee.

Spotify is so well-known, I doubt they need any "prominent" display or Apple driving traffic to them. They'd be quite happy if their app is available on the App Store when requested / searched for and they didn't have to pay commissions or weren't prohibited from advertising on their app.

No doubt Spotify would like access to all of Apple's customers for $99/year. I suspect Apple will find other ways to get paid, from hosting fees, d/l fees, etc. to make up for lost revenue. On advantage Apple has it it knows what the revenue stream was and can set new fees to recoup it. Once Spotify is free to advertise in app and self host, arguments about unfairness go away and they will live in a new world that may not be as nice as they thought.

 

[arobert3434]

They can do what they want in their own app.

They sure can. And I'd imagine the lack of complaints has a simple explanation: the affected parties are individuals, not billion-dollar companies like Fortnite and Spotify.

 

[Abazigal]

Your $ calculations seem way too high. How much do you figure it costs Apple to run/maintain the App Store per year?

I tried to do some calculations last year, but since nobody except Apple knows the exact numbers, it's really an educated estimate at best.

Minnesota and Arizona Bills Aim to Let Developers Skirt Apple's In-App Purchase Rules

Here's my thinking. App Store profits is derived from two areas. The cost of running the App Store (which doesn't change, because the number of developers is assumed to be fairly constant), and the revenue from games and IAPs (which is variable in nature). We also know that about 85% of...

forums.macrumors.com

My napkin math concludes that Apple needs about 20% App Store cut to break even. Detailed breakdown and explanations are in the post linked above. While this is almost from 2 years ago, I don't expect the breakeven number to have changed too much over time. You can plus/minus 1-2%, bottom line is that reducing Apple's cut to 15% likely isn't enough (remember that Epic charges 12% and still wouldn't be profitable if not for Fortnite profits).

Either that or raise the annual developer fee to $1000, which would just cover the cost of running the App Store. While also chasing away every other indie developer.

 

[jlc1978]

Your $ calculations seem way too high. How much do you figure it costs Apple on average to run/maintain the App Store per year?

Even if they are high estimates there a real costs to Apple hosting an app and not getting paid for it; and no good reason to force them to host for free.

If you’d make me name „fair“ number for Apple‘s commission rate, I‘d say it may actually be 10-15% or close to that (if you look at alternative full-service options like FastSpring, Paddle or Gumroad that also do VAT and act as merchant of record, charging somewhat between 5 and 13%. And also factoring in a premium for Apple as the platform operator, their marketing of third-party apps and their app review process).

However, do those service also host the app, and provide all the other services Apple does? Those costs need to be factored in as well to arrive at what a developer would pay outside of Apple.

Capitalism benefits people when there is an open and competitive environment. When regulation, patents and other obstacles prevent new entry into a market then you end up with a duopoly or monopoly controlling everything. Preventing one company from controlling everything is keeping capitalism alive and healthy.

The impact of monopolies on consumers is interesting and somewhat ambiguous. Even Standard Oil's monopoly, the text book case, reduced the price of kerosene from .26c/gal to .07c/gal; which benefited consumers due to lower prices.

In other cases, monopolies harm consumers; such as the old company towns and stores.

My napkin math concludes that Apple needs about 20% App Store cut to break even. Detailed breakdown and explanations are in the post linked above.

And a 10% margin on revenue is not reasonable. Is suspect many on MR do not remember the old pre App Store days when bringing a product to market was expensive and a developer was lucky to see 30% after all costs were taken into account and unless you had the cash or were willing to bet on shareware you could not even get into a market; let alone one as large as the App Store's.

Either that or raise the annual developer fee to $1000, which would just cover the cost of running the App Store. While also chasing away every other indie developer.

I suspect the indie developer would be the most hurt by changes in fees. Th big ones like Epic and Spotify can handle them, but larger upfront and ongoing fees uncoupled from actual revenue would hurt smaller ones without the cash to stay in business.

Edit: Typo fix

 

[arobert3434]

The App Store actually display a current size of 179 MB for Spotify.
But you're calculating it wrong. Very wrong.

Here's an example of a 47MB app: https://apps.apple.com/app/ia-writer/id775737172
Yet when I update that through the App Store, it needs to download a update package of merely 593KB to update the last to the latest version:

View attachment 2325009

Thanks to the clever use of app update packages (that you may want to read up on), the update size is reduced to - in this case - a small fraction of less than 1.5% of the full download.

This was the first and random example showing up in my App Store updates, and admittedly, this must have been a small code update without updates to other app resources (images). But Organic Maps is also displaying 623KB update on a 115MB full download app - so not an extreme outlier.

EDIT:

60 Petabytes is approximately 60 million Gigabytes.
Your $3 million per month figures assumes a cost of $0.05 per Gigabyte.
Some quick googling convinces me that it actually must be less than one cent per GB, at Apple's size.

Given your assumptions about number of users and app downloads and mine about size, that'd make for less than $10k a month (on small updates).

👉🏻 That’s a bit less than Tim Cook’s 2022 hourly executive compensation (almost $100 million for the year. Or more than 10k per living hour - basically assuming the guy doesn’t sleep but works 24h a day).

Again, that’s just assuming small code updates - but anyway, it’s a fraction of your estimate.

I just checked using your method for the following apps: AirBnB (121MB), LinkedIn (73MB), Uber (86MB), Google Translate (29MB), and Reddit (44MB). I also have iA Writer though and got close to what you got there.

So maybe the number for Spotify (don't have it on my phone to check) is in the hundreds of thousands per month not millions, but his point stands. Apple faces real costs for hosting these apps so it's a tough call to say they should get nothing for it and make it all up in device sales. On the other hand a pretty good deal for the multi-billion dollar app company that can put distribution costs as a 0 in their accounting books.

 

[Abazigal]

In other cases, monopolies harm consumers; such as the old company towns and stores.

There's another reason why I feel current antitrust laws don't really apple to modern companies like Apple. OG monopolies harmed consumers by controlling supply. For example, Corporation X is the only source of a particular type of life-saving medication in the vicinity, and so they are free to jack up the price and consumers have no choice but to pay because they have nowhere else to go. In this case, the impact to consumers is pretty clear-cut.

That's not the case with companies like Apple, Google and Amazon, where consumers are attracted to them because they do deliver a genuinely superior experience. This attracts modular suppliers (different people all selling very similar stuff), which improves the experience and thus attracts more consumers, and thus more suppliers in the aforementioned virtuous cycle. The implication is that eventually, one company ends up growing so big by virtue of having captured most of the consumers and most of the suppliers. The difference between these aggregators and traditional monopolies is that consumers self-opt into those platforms because it's a better overall experience. This has completed neutered US antitrust legislation, which looks at harm to the consumer, usually in the form of higher prices or fewer choices.

In the case of Apple, the App Store, by virtue of having aggregated the best customers in the world, has drawn a ton of developers to the platform. The result is a glut of apps available, which has served to drive the price down to zero due to excessive competition amongst developers. So it's hard to make a case that the consumer is actively being harmed here, when the problem is the opposite - there are now way too many apps (to the point where many people complain of not being able to readily locate the apps they want), at extremely affordable prices (users expect a $1 app to be maintained indefinitely).

And I can argue that developers are not Apple's customers, so harm to them doesn't really fall under US anti-trust law, and it's not so straightforward to prove a direct link between a 30% and higher app pricing (which is set by the developer).

 

[webkit]

I tried to do some calculations last year, but since nobody except Apple knows the exact numbers, it's really an educated estimate at best.

Minnesota and Arizona Bills Aim to Let Developers Skirt Apple's In-App Purchase Rules

Here's my thinking. App Store profits is derived from two areas. The cost of running the App Store (which doesn't change, because the number of developers is assumed to be fairly constant), and the revenue from games and IAPs (which is variable in nature). We also know that about 85% of...

forums.macrumors.com

My napkin math concludes that Apple needs about 20% App Store cut to break even. Detailed breakdown and explanations are in the post linked above. While this is almost from 2 years ago, I don't expect the breakeven number to have changed too much over time. You can plus/minus 1-2%, bottom line is that reducing Apple's cut to 15% likely isn't enough (remember that Epic charges 12% and still wouldn't be profitable if not for Fortnite profits).

Either that or raise the annual developer fee to $1000, which would just cover the cost of running the App Store. While also chasing away every other indie developer.

My Apple App Store overhead guess has been that it is around $20 billion per year but maybe it's even less than that. Using my $20 billion figure and assuming there are approximately 1.8 million apps in the App Store, Apple’s average cost per app per year would be a little over $11,000.

The other poster (truthsteve) stated that, "That's 60 Petabytes every month from just Spotify. Based on standard S3 pricing, that's $3 million per month. Spotify no longer has inapp purchases so Apple loses $3 million/month from one app."

I find it hard to believe that if the per app average cost is around $11,000/year that the Spotify app would cost Apple $36 million/year ($3 million/month times 12) or more. Spotify would be higher than the average, yes, but $36 million higher? That’s why I questioned the numbers.

 

[webkit]

So maybe the number for Spotify (don't have it on my phone to check) is in the hundreds of thousands per month not millions, but his point stands. Apple faces real costs for hosting these apps so it's a tough call to say they should get nothing for it and make it all up in device sales.

Based on my calculations in post #323, Apple's App Store average overhead cost per app per year would be a little over $11,000. Therefore, wouldn't even hundreds of thousands per month for Spotify be too high?

 

[jlc1978]

Based on my calculations in post #323, Apple's App Store average overhead cost per app per year would be a little over $11,000.

One problem with averages is the distribution of costs across apps, based on volume, for example, is probably wide with some apps costing a lot more and others virtually nothing in comparison. An average hides the actual costs for individual apps and thus doesn't accurately represent Apple's costs.

Therefore, wouldn't even hundreds of thousands per month for Spotify be too high?

The cost of a product has little bearing on price beyond that needed to be profitable. If access to Apple's user base is worth millions a month to Spotify the Apple should charge them millions. You price at the level the market will bear that supports your financial goals.

That's not the case with companies like Apple, Google and Amazon, where consumers are attracted to them because they do deliver a genuinely superior experience. This attracts modular suppliers (different people all selling very similar stuff), which improves the experience and thus attracts more consumers, and thus more suppliers in the aforementioned virtuous cycle. The implication is that eventually, one company ends up growing so big by virtue of having captured most of the consumers and most of the suppliers. The difference between these aggregators and traditional monopolies is that consumers self-opt into those platforms because it's a better overall experience. This has completed neutered US antitrust legislation, which looks at harm to the consumer, usually in the form of higher prices or fewer choices.

Technology has amplified the impact of network externalities on a business' growth. Monopolies traditional grew by acquiring competitors or cutting prices to drive them out and thus consolidate as an organizational activity. Network externalities mean as you grow you become more desirable, as you point out, and eventually become a dominant force.

Network externalities drove the creation of the so-called natural monopolies in telephone and power deliver as having one standard way to interconnect and deliver services was viewed as the most beneficial and cost effective way to provide those services. No one wanted 10 different phone lines simply to be able to call everyone you wanted to call, or a bunch of competing power lines just to get power.

Similarly, having a choice of apps that is large and silly being able to text and share with friends drives the development of Apple and Google as the two winners in the market.

 

[webkit]

One problem with averages is the distribution of costs across apps, based on volume, for example, is probably wide with some apps costing a lot more and others virtually nothing in comparison. An average hides the actual costs for individual apps and thus doesn't accurately represent Apple's costs.

I noted that Spotify would be higher. My doubt, however, is that it would be $36 million per year higher than the average. That seems way too much of a difference and why I questioned the numbers.

The cost of a product has little bearing on price beyond that needed to be profitable. If access to Apple's user base is worth millions a month to Spotify the Apple should charge them millions. You price at the level the market will bear that supports your financial goals.

This was not about the "worth" to Spotify (which would be a valid separate discussion) but rather the actual overhead cost to Apple. I just don’t see how that cost could be $36 million/year or more for Spotify when the average (based on my calculations using $20 billion total cost per year and 1.8 million hosted apps) is only a little over $11,000.

 

[Flynnsworth]

read the part before that: "but either way, yes. professional services cost a referral fee which is a percentage of the service cost."

do you admit you were wrong now? I await for your response.

No. I think you’re bending over backwards trying to defend anticompetitive practices which ultimately hurt the consumer. Apple owe you nothing.

 

[jlc1978]

I noted that Spotify would be higher. My doubt, however, is that it would be $36 million per year higher than the average. That seems way too much of a difference and why I questioned the numbers.

Fair enough, especially since Apple doesn't break out numbers in a reasonable way.

This was not about the "worth" to Spotify (which would be a valid separate discussion)

True, and what Apple could and should charge Spotify is based on what access to the App Store is worth to Spotify and not what it costs Apple in overhead.

but rather the actual overhead cost to Apple. I just don’t see how that cost could be $36 million/year or more for Spotify when the average (based on my calculations using $20 billion total cost per year and 1.8 million hosted apps) is only a little over $11,000.

It depends on how you allocate overhead. If it is by d/l volume or number of users then Spotify would likely be allocated a lot more than $11K.

For example, if the top 10 apps by user base has 50% of the total user base, then you could allocate 10 billion over 10 apps or 1 billion each.

Doing it by % of revenue would yield a different number.

It comes down to how you do the cost accounting.

 

[truthsteve]

Your $ calculations seem way too high. How much do you figure it costs Apple to run/maintain the App Store per year?

much more than it receives from $99/year developer program subscriptions.

 

[truthsteve]

No. I think you’re bending over backwards trying to defend anticompetitive practices which ultimately hurt the consumer. Apple owe you nothing.

you ran out of arguments considering you can't back up the statement " The only services that Amazon list are the prepaid gift card type." and resorting to insults.

we're done here.

 

[Flynnsworth]

Glad you’ve admitted defeat. All the best

 

[AppliedMicro]

Disingenuous example

It was a counterexample.

Disingenous is claiming Spotify would update a 150 MB in their app four times as month to every user - when as the developer you claim to be you should know better. Even your own example (Tesla) is just a fraction of the full download. Of course there's other data transfer - but on-demand resources download? Come on, that's even in Apple's documentation only used for very large apps (e.g. games).Spotify loads its content from their own servers or contracted CDN. 60 Petabytes is nowhere "conservative".

Why are you using providers where there is no evidence to show Apple is using those providers?

Apple, at their size, will pay much, much closer to marginal cost. Low-cost providers are a good indicator for marginal cost.

Of course there's other data transfer - but on-demand resources download? Come on, that's even in Apple's documentation only used for very large apps (e.g. games).

No doubt Spotify would like access to all of Apple's customers for $99/year. I suspect Apple will find other ways to get paid, from hosting fees, d/l fees, etc. to make up for lost revenue

That's why legislators need to mandate allowing sideloading.
Spotify can host their app themsevels - that's it.

However, do those service also host the app, and provide all the other services Apple does? Those costs need to be factored in as well to arrive at what a developer would pay outside of Apple.

It's literally a few cents per month and user - even in truthsteve's example.

My napkin math concludes that Apple needs about 20% App Store cut to break even

That's way, way too high.

"There are 1.8mn apps in Apple’s store and, although more than 80 per cent of those are free, the judge in the 2021 case agreed with an estimate that its operating margin exceeds 70 per cent."

Google and Apple’s app stores are cash machines

This week’s antitrust victory for Epic Games may disrupt two lucrative digital outlets

www.ft.com

 

[truthsteve]

It was a counterexample.

Disingenous is claiming Spotify would update a 150 MB in their app four times as month to every user - when as the developer you claim to be you should know better.

Like I said, dumbing it down + factoring in many users have more than just 1 apple device + push notifications + on demand resource downloads means 150MB isn't too far off. But even if it was far off, the point still stands, $99 covers about 1TB of bandwidth.

Apple sells hundreds of millions of iPhones and iPads per year. Spotify being installed fresh on a fraction of those new devices exceeds 1TB.

Even your own example (Tesla) is just a fraction of the full download.

Disingenuous again, Tesla's update equated to 40% of full download. That's not what I would characterize as "just a fraction". That's a "substantial fraction".

Of course there's other data transfer - but on-demand resources download? Come on, that's even in Apple's documentation only used for very large apps (e.g. games).

Nope. Wrong. It is not *only* used for very large apps. The company I worked for used it for one-time background videos during the signup screen. That way returning users can immediately login faster due to smaller app bundles needed to be download while new users get an elegant sign up screen. And it doesn't cost us extra. Our app was < 100MB for initial release.

Spotify loads its content from their own servers or contracted CDN.

Songs and album art? Sure.

60 Petabytes is nowhere "conservative".

You completely ignored the part about users having multiple apple devices and upgrade cycles. Congrats.

Apple, at their size, will pay much, much closer to marginal cost. Low-cost providers are a good indicator for marginal cost.

LOL, and you think $99 is enough to cover all of Spotify's data transfer from the App Store?

Sorry, but you are literally wrong. Even if you are correct, what about App Store Review? 50 updates a year = 50 app reviews = 50 x 30 minutes = 25 hours of App Store reviews. App Store reviewer makes, what? $20/hr? That's $500/year Apple spends just to review Spotify.

Tell me again how $99/year covers Spotify.

Of course there's other data transfer - but on-demand resources download? Come on, that's even in Apple's documentation only used for very large apps (e.g. games).

Straight from the docs

Remote storage of rarely used resources. The app has resources that are used infrequently. The resources are requested as they are needed. For example, an app tutorial is usually shown once after the app is opened for the first time, and may never be used again. The app requests the tutorial on first launch, and then requests the tutorial only when needed or when new features are added.
Remote storage of in-app purchase resources. The app offers in-app purchases that includes additional resources. The resources for purchased modules are requested by the app after it is launched. For example, a user purchases the SuperGeeky emoticon pack in a keyboard app. The app requests the pack after it finishes launching.

No where does it say *only* for large apps. It literally showed a suggestion to where we implemented ODR in our <100MB app and suggested it's for a keyboard app which I don't know about you, but I wouldn't consider a "keyboard app" "very large".

You're wrong *again*.

 

[AppliedMicro]

and you think $99 is enough to cover all of Spotify's data transfer from the App Store?

Sorry, but you are literally wrong

Calling me „literally wrong“ for something I never even remotely claimed is disingenuous.

There’s no doubt that $99 a year doesn’t cover Apple‘s cost of hosting and delivering the Spotify app for all of their users.

It is not *only* used for very large apps. The company I worked for used it for one-time background videos during the signup screen

So it’s for a comparatively large (video) asset that’s irrelevant to the app‘s core functionality.
You didn‘t change that 4 times a month with every update, did you?

what about App Store Review? 50 updates a year = 50 app reviews = 50 x 30 minutes = 25 hours of App Store reviews

It will only be a fraction of the 30 minutes you claim for each update (or are you „dumbing down“ again)?

„Most reviewers spend only a few minutes per app, but many apps are simple and only require a short period to evaluate, the people said.

Reviewers have daily quotas of between 50 and 100 apps“

Inside Apple's team that greenlights iPhone apps for the App Store

An executive board led by Apple marketing SVP Phil Schiller meets every week to discuss controversial apps or other iPhone software programs that may infringe Apple's App Store guidelines.

www.cnbc.com

👉 At 50-100 apps a day, the reviewer does not take 30 minutes on every tiny iteration of a long-standing and well-reputed app such as Spotify. Unless they’re making an exception for Spotify - which they wouldn’t need to, given Spotify‘s proven track record of obeying Apple‘s current rules (irrespective of how they may be trying to change them through public lobbying efforts and possibly legal action).

 

[truthsteve]

There’s no doubt that $99 a year doesn’t cover Apple‘s cost of hosting and delivering the Spotify app for all of their users.

And raising it to millions per year would be stupid.

So it’s for a comparatively large (video) asset that’s irrelevant to the app‘s core functionality.
You didn‘t change that 4 times a month with every update, did you?

suddenly *only* "very large" apps

changed to

"comparatively large" apps

moving goal posts, huh?

FYI: the movie was about 12 MB seasonally swapped out. App with video was still less than the total size of Spotify.

It will only be a fraction of the 30 minutes you claim for each update (or are you „dumbing down“ again)?

1. 50 updates a year would be approved updates. Rejections happen a lot and requires resubmission. 25 hours is conservative.
2. "but many apps are simple" Spotify is not simple
3. App reviewers don't receive a list of changes, so an app with many screens can't reasonably be checked in a few minutes.

not dumbing it down this time. 30 minutes is realistic.

👉 At 50-100 apps a day, the reviewer does not take 30 minutes on every tiny iteration of a long-standing and well-reputed app such as Spotify. Unless they’re making an exception for Spotify - which they wouldn’t need to, given

Again Spotify isn't a simple app. Are there a lot of fart apps? Sure. That makes up the majority of 50-100 apps. Spotify is not a simple fart app.

Spotify‘s proven track record of obeying Apple‘s current rules (irrespective of how they may be trying to change them through public lobbying efforts and possibly legal action).

Sorry, you're wrong *again*.

" Apple has reportedly rejected Spotify's latest app update three times in the last month."

https://www.macrumors.com/2022/10/25/spotify-audio-book-app-update-rejected-apple/

 

[scorpio vega]

Not me forcing my rommates/brother/dad to delete spotify and take advantage of Apple music though
(Best buy has 3 month trial lol)

 

[AppliedMicro]

not dumbing it down this time. 30 minutes is realistic.

It's not realistic for the average app.

I said it before: "Unless they’re making an exception for Spotify" by giving it special scrutiny...

" Apple has reportedly rejected Spotify's latest app update three times in the last month."

https://www.macrumors.com/2022/10/25/spotify-audio-book-app-update-rejected-apple/

Quote: "Apple initially approved the update in September, but later reversed course, rejecting subsequent updates."

Initial approval does not prove they're taking much time with every update. Quite the contrary: it casts doubt on whether they're spending much time on every update. Especially since Apple knew exactly what to look for, as Spotify was known to enter the audiobook market.

Also, this is a particular focus for Apple as part of their anticompetitive anti-steering policy. As the article says: First they're allowing some form of external information to appease regulators - before subsequently weaselling their way out of actual compliance.

"Apple last year agreed to an App Store rule change that allows developers to use communications like email to share information methods about payment options that are available outside of an iOS app, but Spotify is running into issues attempting to implement this feature"

In this case, Apple's reason for rejection has anyway been found illegal not only by U.S. courts twice now...

"Apple has been ordered to implement ‌App Store‌ changes that will allow developers to use metadata, buttons, links, and other calls to action to direct customers to purchasing mechanisms outside of the ‌App Store‌, paving the way for developers to implement alternate payment methods."

Judge Lets Apple Further Delay Implementing App Store Rule Changes From Ongoing Epic Dispute

Apple does not yet have to implement changes to its "anti-steering" App Store rules that would allow developers to direct customers to...

www.macrumors.com

...and will probably also be found illegal (according to existing competition law, not the new DMA) by EU regulators - exactly what the original Macrumors.com article here is all about.

 

[jlc1978]

That's why legislators need to mandate allowing sideloading.
Spotify can host their app themsevels - that's it.

Certainly; just don't expect Apple to host it for free as well.

It's literally a few cents per month and user - even in truthsteve's example.

I'm not so sure tax and regulatory compliance would be a few cents per user for most small developers. Spotify and Apple, probably as costs are spread over a large user base. Compliance costs have a fixed and variable component for taxes, for example, and it's teh fixed that's likely to drive the costs for smaller developers.

Then there is the costs of chargebacks and the risk of losing the payment processor if they are too many chargebacks.

My suspicion is small developers will not see much advantage from side loading and going it alone due to the cost; especially those Apple only charges 15% commission.

And raising it to millions per year would be stupid.

Not necessarily. Apple's choice would be to host Spotify for $99/year and make very little in commissions, or charge Spotify new fees to make up for lost revenue.

Apple needn't do that right away. They know how much they make from Spotify now, and can see what the impact is of Spotify being able to direct purchasers away from IAP. If the loss is too big, they can come up with new fees. If Spotify leaves then Apple is no worse off if the drop was already impacting revenue significantly.

If Spotify leaves before side loading becomes universal, they lose their iOS app and are back to a web version, plus they need to convert all the subscribers, see off whom may not want to lose the IAP option.

The bet is Spotify may need the App Store more than Apple needs Spotify. Given Apple's financial position vs. Spotify's, I'm guessing they can outlast them if they want to.

 

[webkit]

It depends on how you allocate overhead. If it is by d/l volume or number of users then Spotify would likely be allocated a lot more than $11K.

For example, if the top 10 apps by user base has 50% of the total user base, then you could allocate 10 billion over 10 apps or 1 billion each.

Doing it by % of revenue would yield a different number.

It comes down to how you do the cost accounting.

I don't doubt Spotify would be more than $11,000. I was questioning the $36+ million.