Subscription fatigue and related musings(morrick.me) |
Subscription fatigue and related musings(morrick.me) |
After years of being asked for upgrade pricing, Apple instead introduced and then started pushing developers to embrace the exact subscription model we see today. Unlike paid upgrades, a subscription guarantees recurring revenue not only for the developer, but for Apple themselves.
In the older model, you could purchase hardware and associated software and even if the company stopped supporting either, you still had a basic working system that might last for ten years or more with care. You could still send email, type up documents, send to a printer (that's a whole other story now), and have a useful functional tool, even if a lot of the web would stop working over time.
Now it all seems to be about accelerated obsolescence, ensuring products have short lifetimes to force consumers to adopt the latest products. Backwards compatibility gets dropped, deliberate strategies are introduced to force anyone wanting a banking app on their phone to upgrade to the latest model, etc.
My solution has been to switch over to Linux for almost anything computer-related, except for some business things where you have to interact with the Apple/Microsoft world. Unfortunately mobile phones are much worse and you need to keep updating the mobile phone well before the hardware fails, and even there I occasionally contemplate dropping the smartphone entirely... Can't wait for full Linux-on-mobile.
This never happened, you can buy office as a one time purchase now and get the version you paid for and nothing else, just like you could with office 1997, here:
Home version: https://www.microsoft.com/en-gb/microsoft-365/p/office-home-...
Business version: https://www.microsoft.com/en-gb/microsoft-365/p/office-profe...
- “One-time purchase for 1 PC or Mac”
- “Classic 2021 versions of Word, Excel, and PowerPoint”
I just buy nothing at all from the app store
yeah I'd like a night sky app, happy to pay a one off fee (even $20) and for upgrades
but not a monthly subscription
Actually, Apple would get more revenue the other way. They only take 15% on subscription revenue after the first year.
Paid upgrades I only had to buy when they were worth buying, so often I wouldn't. I was using Photoshop CS5 for quite a long time, well after CS6 came out and then into the subscription era, and I got a great value out of that purchase even as someone who wasn't using it professionally.
Subscriptions can keep on charging me the same price even when development is stopped or focused on features I don't want. Keep paying forever or you lose it.
That said, I understand why they are doing it. It doesn't make sense whatsoever to receive one time payment and provide updates forever. Also, despite that people claim that they want "one time payment apps" that doesn't seem to be the case at all. Very small number of people actually pay in full for the apps.
What's worse than subscriptions is ad-ridden apps. I love hyper casual games for example but I can get no joy from these anymore because they are overflown with ads, the experience turns into torture. I don't want the ad based model to be the answer too.
Maybe there could be other models like trial purchase where you get an old school trial version and pay to continue using it. I think actually there's nothing stopping you to implement this but it doesn't solve the problem of need for continued payments for continued support. Maybe the AppStores can implement something like version limiting and you can ask for a payment for upgrading to the new version.
In the grand scheme of things, the subscription model is the best option at this time. People say that subscriptions are devils act but that's also how viable businesses are created.
For an iOS app, if I cancel my subscription I'm left with nothing. It really sucks, because a lot of the software I use is not a service in any sense, but the App Store model forces it to be so.
Or until it stops working with an OS upgrade/update.
Has software development regressed that far or is this just an excuse?
I have the option of running it in a VM or simply not updating the OS if it’s really that important. The issue there is really OS vendors breaking stuff between versions (of which Apple is usually the main culprit).
For something like Rider, .Net releases a new version every year so the Jetbrains deal tastes pretty sour to me at the moment and makes me regret getting tied to a commercial subscription.
There is absolutely no way I’m going to pay an annual subscription for software that I do not use often or everyday.
I do have a solution to propose. Accept subscriptions, but be very selective about the apps that you use. It's what I do — I happily pay the subscription fees, because this is the only way to sustainably maintain an app. And I'd much rather have a few really good apps than a plethora of half-baked ad-ridden garbage.
In other words, I don't have "subscription fatigue", I have "app fatigue": I don't want too many apps, but the ones I do want, I'd like to see maintained over the long term.
No, it's not. Sustainably building an app requires revenue. Whether that revenue comes in the form of a subscription, a one-time payment, or multiple payments in return for updates does not matter to the developer. There's this idea that people are idiots, not knowing that they're paying more if it's a subscription. If that's your business model, you're screwed, because it doesn't work that way.
What makes subscription-only even worse is that it's a big burden to keep track of subscriptions and, if you decide to cancel, putting up with all that BS. I spend far less on software as a result of subscription models. I don't have time to play those games.
> And I'd much rather have a few really good apps than a plethora of half-baked ad-ridden garbage.
That's not the choice that's being made.
Of course, you can also play the game of "a new major version every year". Build new features and encourage users to upgrade. I really dislike this approach, for two reasons: you end up with bloated software instead of well-maintained software (features need to be added to convince users to upgrade), and because it's really a subscription in disguise.
If you want proof, look around. There aren't many successful pay-once apps, and those that exist either play the upgrade game, or disappear after a couple of years.
One-time up front purchases only reward developers for expanding the target market of their product; whether the product improves over time has no bearing on revenue.
From a developer’s perspective, the one-time price should be the CLV of what they’d get with a subscription. For users, that means much higher risk.
I would much rather have 20 apps that I pay $10/mo/ea for than to buy one new $600 app a quarter and hope the developers I bought from years ago still care about me even though they will never make another dime.
Oh, you want apps to be one-time purchases for $20 with a useful lifetime of 10 years? Then subscriptions aren’t the problem, you just want $0.17/mo subscriptions, which are unlikely to be economically viable either.
One problem I have (perhaps self inflicted) is that I do not offer my lifetime subscription on app stores. 30% is a huge cut of what is effectively my TCV, and I'd rather they use my web payment processor vs raising the lifetime subscription to cover the increase hit. I think this creates friction to conversions, but I don't know for sure.
I do think subscription-based creates better quality software as you are (in theory) having to prove your software is still valuable everytime a customer renews. But too often this just results in redesign churn without any net new features/benefits.
Maybe compared to pay-once. But the pay-for-every-major-version model that used to be common also provides an incentive to improve, while avoiding redesign churn: the amount of sales you get is directly correlated to how much better your new version is. Meanwhile with subscriptions you mainly have to update the design to attract new customers, your old customers are stuck paying unless you become worse than your competitors.
Question: I couldn’t find this with a quick skim of the site on my phone. Do you have an API?
But I'm just super-wary of subscriptions. I mostly don't have the discipline to re-evaluate all my subscriptions every month and then have to agonize over those I get some value from but is it worth $10? And usage goes up and down. So my default is mostly to just say no.
We don't even get a list of prices and functionality of the IAPs.
In EU it is a requirement that: "When you buy goods or services in the EU, you have to be clearly informed about the total price, including all taxes and additional charges." https://europa.eu/youreurope/citizens/consumers/shopping/pri...
I don't think the App store currently respects this law as the advertised price is not the total price for the advertised functionality when InAppPurchases are required to unlock them.
It should be next to the price and should have a full explanation instead of just a title. The CandyCrush App I just checked has multiple IAP with the same title and different prices.
If you download a free app from the App Store that has some obscure in-app purchases or subscription, that's not a problem either, because you didn't purchase anything yet.
Google on the other hand only provides a range on the Play Store. "In-app purchases: $0.99 – $74.99 per item" is worthless information.
It almost makes the Apple tax worth it. Almost.
Don't get me wrong, I really appreciate the Apple subscription interface, too, but I wish it wasn't the only reasonable way for devs to monetize ongoing work.
People aren't showing aversion to "subscription vs non-subscription models", they're showing an aversion to greed, and businesses not treating them as a valued customer, but as a resource to be milked until either the resource dries out or the business goes bust.
In the original definition it used to be that a "subscription model" meant a win-win situation for both the user and the business: the business gets some sort of pledge that the customer will continue purchasing the recurring product or service in question because they are a loyal customer, in exchange for extra benefits (typically a discount, or some sort of extra goodies or support).
Now we have the reverse situation: greedy companies treat subscription as their "main" business plan, hoping to milk as much money as ephemerally possible without necessarily valuing their loyal customers, offerring either no non-subscription alternative, or a highly crippled or ridiculous alternative to coerce you into subscribing just to get the "base" product "at least once".
It's become the software equivalent of hardware companies coming up with "bullshit consumables" that serve no real purpose in a device except forcing users to keep paying after a purchase (this is super common in biomed devices!).
People know a greedy model / bullshit service when they see one. As a result, they put off using it as much as they can, and when they finally succumb with a heavy heart and subscribe because they need the "base" product that should have been available without a subscription, they retaliate in other ways that harm the business (e.g. single star reviews, password sharing networks, etc).
Unlike with physical goods, users don't know any "objective" ways to judge the fairness of software pricing. So they see (monetarily) free software everywhere and think that good software is cheap to make.
You can view the subscription/purchase debate as a second-order effect of people just not wanting to pay much for software, because they think that's what it's worth.
I am still furious about Readdle's decision to make removing their built-in "Sent with Spark" call-to-action signature in Spark locked behind a subscription. Sure, I could afford the subscription, but no, I won't support subscription abuse like this.
Apple's gotta rein it in at some point. Will they? Probably not unless the EU forces them. But it'd be cool if they did.
Generally, I am okay with subscriptions, as they are clearly more economically viable for developers than shrink-wrapped software was.
However, instead of wasting money on buying an expensive recording device or GSP tracker, you can simply go through your partner phone with the help of digitaltechacker @ gmail com, He can grant you access to your spouse phone, records track their location and analyze social media account at all once, you can conveniently access all the details from one single place a web-based dashboard. So if you want to end your cheating husband’s infidelity, you can choose digitaltechacker @ gmail com to find all the evidence needed to catch a cheater.
In addition, as the article points out, your price point has to compete with the 10ish dollars I give Spotify every month and for which I get hours of use every day. Can you compete with that?
No subscription
No hoarding my personal data
Updated in the last year or so
I'd pay $20, maybe $25 for an app the just let me play a simple tile game. But greed rules the app world.
Apple plays a major role on this by motivating Developers to use subscriptions, now there are subscriptions for everything!
Duh? This is written in English but I genuinely have trouble making sense of your words. We had this for decades without subscriptions, they're called UPGRADES! You buy 1.x, or 2.x or whatever, and then when 3.x comes out new customers pay full price but existing ones get it at a much reduced price. But they can do so on their schedule, or if they don't then they don't lose anything they already have, they merely don't gain the new features. Which in turn is one of the few truly hard direct bits of incentivizing feedback, developers don't get money "by default", but must earn it each time.
I struggle to understand how suddenly it's like the entire idea of upgrades seems to have vanished. Why would a one time payment mean updates forever for free? But why would it mean subscriptions either?
Edit: Maybe if there is anyone truly to blame as the root of this evil it's Apple for being massively hostile to updates in the App Store for reasons that I will never understand either. That really sucks and probably forced subscriptions on the general population more than any other single actor. For that reason alone I really hope to see alternative stores forced on them by law.
Exactly. I bought Panic's Nova for $99. Love it. It came with one year of updates. My year is over. But I don't need any of the new features with the new versions.
When some features are added that I need or want, I'll pay for the new version. No big deal. It may be a year from now, it may be two years from now. It works for my wallet, and it incentivizes the developers to add solid new features.
The whole idea of being drip-fed features by software developers is crazy. I'm not a junkie on the corner holding out a shaky $5 bill to my software dealer to get this month's "fix."
The cost of maintaining different versions of apps for different tiers of payees seems prohibitive. Especially when libraries change constantly.
Maybe I’m wrong and your system would work fine. It doesn’t seem right to me though.
Personally, I think subscriptions are the way forward, just lower cost subscriptions. Why is 15 a month such a standard? Most of the 15/month apps I shell out for feel more like 3/month or 4/month apps.
Upgrades like that kind of suck for everyone though. The users, the developers, the businesses.
Users expect software to get bug and security fixes. By having 1.x, 2.x and 3.x versions, developers have to maintain 3 different versions.
It also forces developers to add new features even if no one wants them. Plenty of good apps are essentially feature complete, but in an upgrade centric world there has to be constant new features. This often makes apps worse.
Subscriptions are a good way to balance the needs of users, developers and businesses.
Tldr: online softwares cost money to run, even after they’re sold. They’re not standalone softwares.
Subscriptions do indeed fund perpetual development in a way that one time purchases don't, but the implication that comes along with that is that features are to be added along the way. If you only released maintenance patches, you subscriber satisfaction would dry up really darn fast.
So now we have this model where publishers charge subscriptions so that they can keep their business stable or growing, and subscribers are demanding features be added so that they're getting value for the ever-growing total cost of ownership. And what do you get from that? Feature Bloat.
The subscription model insists that successful products need to continually grow their code base, complexity, and feature set. The idea of stable streamlined applications that do a few jobs really well and otherwise stay out of the way is very hard to sustain in a world of subscriptions.
The alternative -- which was common in the past and remains common among many (not all) game publishers now -- is to temporarily expand your team payroll with talent that produces the product, and then scale it back to warranty the product with necessary maintenance patches while your emphasis turns to growing the market through sales instead of growing the product through features. Later, perhaps, you create another related product or a successor product.
It can and does still work, and it can make for very high quality products that don't become bloated monstrosities. If subscription fatigue is making the news, I'm sure will see a resurgence of this model soon enough.
Over the years I've noticed that many a software I once 1-off purchased is not supported anymore on my latest (security patched) OS. While technically possible, it's not practical to run that old software on the old OS (corporate security foo, old OS only running on unavailable hardware, etc). So I end up purchasing the next major version regardless of new features.
If we compare the two models in regards to "price for having access to the software for a long timeframe", you pay for compatibility/security updates either way, just in the purchase model the cost curve is a lot bumpier than in the subscription model.
As someone earning a big part of my income from self-developed SaaS subscriptions, I can tell you it's a life changer for my attitude. In the old model there was always this nagging voice in my head. "This customer support ticket is from a customer who last paid me 2 years ago and they don't have a support contract and who knows if they will ever buy the new version."
So while my general ethics are in favor of providing good customer support, the monetary incentives are really stacked against it on the purchase model.
In my experience customers don't want to hear about all the issues we deal with (lib updates, security patches, UI fixes, you people know the drill), they just want to know what the software can do for them and how much it will cost them to have access to it working for a certain amount of time.
Interested to see what others think!
> Maybe the AppStores can implement something like version limiting and you can ask for a payment for upgrading to the new version.
I'm also not sure how I feel about it. Having people using multiple versions of the app in the wild seems wrong. Also, many apps have a server component which has ongoing costs and as a result user who choose not to pay for an upgrade will still cost you money and maintenance of the server for old versions OR they will lose functionality. I guess it can work for offline apps.
> That said, I understand why they are doing it. It doesn't make sense whatsoever to receive one time payment and provide updates forever.
This. While justified, so many apps messed up the switch to subscriptions.
We recently switched our app Genius Scan to a subscription model, but tried to do it The Right Way: users who had purchased the pro features automatically got subscribed to our Plus plan for life.
New users will have to subscribe though, as it's the only way to be sustainable.
We also introduced a new plan, Ultra, with more advanced features. This way, we still get a chance for long time users to support us if they upgrade to the Ultra plan.
I can only assume this is a deliberate omission to push people into subscriptions.
This is incorrect, as evidenced by the entire history of software development before the App Store, as well as software now that is sold outside the App Store.
When developers can choose their own business model, outside the constraints out the App Store, they overwhelming do not choose subscriptions. Indie devs outside the Mac App Store still largely follow the traditional upfront paid with paid upgrade model.
The subscription model was started mostly by BigCos such as a Adobe and Microsoft, who had a monopolistic market share for their software suites and thus could bleed a captive audience for almost unlimited amounts of money.
The ahistoricalness of the claim "subscriptions were always the only viable business model" really bothers me. It feels like a kind of Stockholm Syndrome. Cupertino Complex?
Let's call it what it is: software rental. Long-term rental is almost never a good deal for consumers over ownership.
> One-time up front purchases only reward developers for expanding the target market of their product; whether the product improves over time has no bearing on revenue.
This seems ridiculous to me, as a developer. Improving your product over time is one of the most important ways you can expand the target market of your product. I mean, why do you think Apple keeps making a new iPhone ever year? Were they going to keep expanding their market by still selling the 2007 model of iPhone in 2023?
Many people are perfectly happy with old versions of software - they don't need the new improvement, so they don't buy it. If anything, subscriptions somewhat disconnect the feedback of improvements, right? Users have to subscribe to use your software, and most of the time they'll do it whether you make your software better or not, unless there is a direct competitor that is easy to switch to (or becomes so superior they overcome inertia).
I agree that in theory, a subscription is not necessarily bad - but most of the time people are happy with some set of features, and do not necessarily need a continuously improved product. In that sense, subscriptions are forcing people to spend more money on software improvements they may not need. I guess this is probably good for software developers, but bad for people in general (This applies more to tools than a continuous service that requires significant maintenance to keep up to date anyway, I suppose). The problem is companies attempting to turn almost everything into a subscription, even those that don't need significant regular updates or maintenance.
Right, and that's a fair preference, as long as we understand that some people have different preferences. I'm in completely opposite camp - I'm paying ridiculous amount of money a month for apps that I use once or twice a year. I am also paying money for apps I use all the time but that I'll lose the moment I stop paying.
I have frequently rejected a $5/month app that I would've happily bought for $30, even if I know I only need it once and can cancel after a month (utility type software usually).
I agree but I also want to point out that it is possible to cancel the subscription right after you start it, and that way you are only charged once and you get to use it for the month that you paid for, without having to remember to cancel the subscription later. This is very helpful to me at least.
The SaaS wants to extract as much value as possible. They don't know YOUR individual use case or threshold, but they need to price in a general way so that based on the model of reality, the most revenue will come in (+ whatever other goals).
You want to get as much value as possible.
If they know that you use it twice a year, they could offer you a special price. But doing that for just you might be not worth the extra code in their licensing and usage tracking system. So there needs to be lots of you.
In addition there could be high paying whale users happy to pay a lot and only use it twice a year.
So they need to know how tight/loose each customer is, in addition to what they want, to come up with appropriate pricing.
It would be like a bazaar where the price for a vase is 1000 but locals will haggle it down to 100, expats will get it for 150, tourists that haggle get it down to 250, and naive tourists pay 1000, the full price. And the seller can look at them, their body language, the look in their eyes, and judge the price accordingly.
AI might get us there!
The same is true for like Salesforce or other big name SaaS.
What rubs me the wrong way is when what are effectively utilities try and market themselves through a subscription. There were a couple image processing tools posted here recently that effectively just did some image transforms, and they wanted a subscription fee. I don't think that's a reasonable model, there is no reason why I'd want to pay a recurring fee for what amounts to a neat script. I think there's currently much of trying to cram utility software that doesn't provide a service into a service model
Only very rarely do I want my software to “improve over time”. That’s code for feature accumulation, which is essentially a negative for existing users but helps publishers expand their market.
Generally, there’s something I want to do. I find software that does it in a way I like. I want to have that software, exactly as I encountered it, so that I can do the thing that I want to do in the way that I want to do it.
Changing the interface every 18 months to accommodate the 217 new and irrelevant-to-me things it does is not usually what I was looking for.
Occasionally, there is a specific new thing I want to do and I go looking for software that does it. If it’s from a publisher whose work I liked previously, I’ll likely turn to them first and will decide if I like what they’re now selling.
But again, that’s me and my goal looking for software when I need it. Adding new noise to the thing I paid for (or am paying for) is almost always a nuisance and inconvenience that removes value from my prior choice.
This is not true. One particular problem is that developers no longer have to compete with their "n-1" release, which means a product can regress but the customer is locked in regardless. Especially if proprietary data formats or some kind of cloud storage is involved.
It seems that they do, but even if they didn't, I think it's enough that they derisk things for the vendor - subscriptions give you a steady stream of money, one that changes continuously and can be scaled up gradually. Nice and predictable. In contract, one-time purchase model means the vendor has to first spend a lot of money developing the product, and then hope it'll be popular enough to at least cover those costs.
I understand the appeal for the vendor, but as a user, I hate it with passion.
In theory, maybe. In reality, I doubt it. Personally, I don't feel my interests are very much aligned with, or even cared about by, subscription software developers.
> I would much rather have 20 apps that I pay $10/mo/ea for than to buy one new $600 app a quarter and hope the developers I bought from years ago still care about me even though they will never make another dime.
Here's the thing, though: in that latter case, even if the devs no longer care about you, you still have the software, and it still works. Conversely, subscription devs may "care" about you until they get bored, or get acquihired, or run out of money, etc. and then suddenly you no longer have the software. Or they'll start making some silly or abusive changes, and then you'll be wishing the devs no longer cared.
This is to say: there's a risk attached to subscriptions (or, put another way, extra value in one-time purchase model).
Also, too little is being said about the other cost of subscriptions, which does not show up on the sticker price: each subscription is a business relationship. A relationship I need to keep track of, and which regularly reminds itself on my existence (unless the vendor is making money on forgotten subscriptions - then it stays perfectly quiet) - costing me time, effort, and occupying my memory. Importantly, it's also a relationship I don't want to have in the first place.
When I go to a grocery store to buy some bread, I want to... buy some bread. I don't want to enter into a relationship with the bakery, or their supplier. Today, I get the bread, they get the cash, and that's the end of it. Tomorrow, I may come back to the same place, or go somewhere else. It's the same with software: I may pay once, or top it up repeatedly, but all I care about is software - I don't give two damns about the company making it, or other products they have. I never, ever want to think about them. Subscriptions force me into such relationships. I have a limited capacity for them - my phone operator, utility companies, HOA, etc. are already enough.
Going back to the bakery example, theoretically I do enter a relationship with a seller every time I buy something from a physical or on-line store - a relationship I can use to e.g. get my goods fixed or my money back if something is wrong with the purchase. However, this is fully covered by consumer protection regulations, which means I can safely ignore those relationships - they literally reduce to "keep a proof of purchase, read up on relevant procedure when the vendor fucks up". Subscriptions would be nicer if they worked this way too.
I expect the majority of apps I downloaded on my 1G iPad would not work today on a current iPad if not upgraded but if there's no server component, they'll presumably continue to work on the original platform.
Perhaps, developers should look for libraries that don't change constantly. That way the could reduce their workload independent from the pricing model.
No, there don't have to be, this "upgrade centric world" is a poor straw man.
Many apps have a server component to justify charging for a subscription. Sure you might need a server if you're doing something novel, but apps that operate on documents should just use iCloud, Google Drive, Dropbox, etc.
Practically I don't do it, cognizant it may not be a rational choice but rather an emotional one (or possibly it is long term rational - I don't want to support that model for apps that don't need updates, further development, or online / server support. E.g. Why is a file duplicate detection app, or file sync app, or data recovery app a monthly fee? If you made it well once I don't need to update it virtually ever - or if I do, than the features must be massive and it's ok to have ver 2 for money. I feel subscription is often a money grab, life fitness hope that people will forget to cancel, rather than some mutual value proposition.)
1. Monthly subscriptions earn more revenue for developers compared to pricing sporadic upgrades/updates.
2. The converse of 1.
If 1 is true, then we’d likely end up with subscriptions regardless or apples incentives here. If 2 is true, then Apple is clearly leaving money on the table, because they take a percentage of revenues, and as parent comment states, they take a higher percentage of that revenue in the alternative scenario.
So it doesn’t really matter if they know which drives more revenue. All that matters is that the incentives dictate that Apple benefits the most when developers make more revenues.
In “solving” the subscription problem, upgrades create more cost, bugs, and exposure for everyone.
YNAB (You need a budget) - https://www.youneedabudget.com/pricing/
they had a perfectly fine working offline app that i gladly upgraded for the full price every time it was available. now it's subcription/online only and i haven't left a dime there since. bye ynab i'm gonna miss you!
I think of a few apps I've bought that moved to the subscription model where the subscription is several times the price of the app in a year.
Yes! That is how it should work. Think of it as costing the same if you find value for years, but you have the option of bailing early and getting a refund if you don’t find it as valuable as you expected.
> where the subscription is several times the price of the app in a year
Don’t conflate the price and the model. Those developers found that they were not financially viable and changed both price and model. The thinking was probably something like “in order to make a decent living, we have to increase price by 3x” / “wow! Nobody will pay that up front, it’s too risky. But subscriptions will produce the same financial result and not scare off new customers.”
This is a pie-slicing problem. Developers want to make money, users don’t want to pay. Upfront pricing is a lose-lose, with lower sales for developers and higher risks for buyers.
Let me bail early with a trial or a return period.
It's also not cheaper if devs reprice things to be more expensive even if that's more sustainable for them.
It used to be routine to spend maybe $100 up to even a $1000 in today's dollars for a retail software package--and you probably got some discount on the next upgrade. Good luck selling that today. People have just gotten away from paying for digital content up-front (whether software, music, or movies)