But in an imperfect world whereby our dependencies ( software components that we use) and platforms that we need to build/rely on ( like Civil 3D) do charge us on annual basis, and that some of users expect perpetual bug fixes from us, with or without a support contract of sorts, SaaS seems to only way to go for our sustainability.
We've all got to push back against these bloated saas models that don't bring tangible benefits to end users and serve only to pad company valuations. Make new versions of your software with features meaningful enough to encourage people to upgrade and outline support periods for existing software sales after they buy a one-time license. There's gotta be a better way. For everyone (except big tech CEOs).
That's why software keep adding bloat fancy buttons and change color scheme every few years. This is anti-productive.
No support contract? Pound sand.
The problem exists from both sides of the coin. Firstly the bulk of customers don't purchase a support contract. So there is very little income to pay staff. So the "support" department has very few people. They're also not very good because low wages means staff turnover.
Then Betty phones with a problem. Significant time is spent explaining to Betty that we can't help her because she (or more accurately her company) doesn't have a contract. She's fighting back because an annual contract seems a lot for this piddly question. Plus to procure the contract will take days (or weeks or months) on her side. And it's not I any budget, making things harder. Betty is very unhappy.
The junior tech doesn't want to be an arsehole and it's a trivial question, and is stuck in the middle.
We switched to a SaaS model in 2011. Users fell over themselves thanking us. They don't have to justify it to procurement. The amount can be budgeted for. No sudden upgrade or support fees. Users get support when they need it. The support department is funded and pays well, resulting in low staff turnover, and consequently better service.
Plus, new sales can stop tomorrow and service continues. Funding for support remains even if sales saturate the market.
Consumers may dislike SaaS, but for business, it absolutely matches their model, provides predictability, and allows for great service, which results in happy Users.
In the companies I've worked for so far since SaaS became a thing you absolutely need to go through procurement for a big enough purchase. You actually need to negotiate the contract each time it expires, which is IMO more burden on the end user than buying a one-off license.
The problem with support contracts, or support requests solved by an upgrade, is that the User needs it now, not after a procurement process.
Doing procurement annually is easier because it can be planned for, budgeted for etc, and happens on a separate thread to the actual support.
Even when they overlap there's enough grace to keep the User happy while waiting on the customer.
What's wrong with SaaS?
If we didn't sell our desktop software to ~1000 companies as a SaaS then few would afford it. We could sell one-off/perpetual licenses for maybe $1M but only our biggest customers would manage that expense, while smaller competitors would not. And if that means we sold only 300 licenses, then the price would be even higher because the number of licenses sold would be even smaller. The SaaS is basically what the customers ask for. They can cancel and switch to competing software when they want to. In fact, customers who use the software rarely feel the SaaS yearly cost is too high so ask for even more SaaS-y functionality such as paying by minute of use or per specific action like "run simulation", instead of having a yearly subscription. Because they might just use it a few days per year so they feel that (say) $10/yr is too much.
That's okay, but in say, 2 years when Mac OS 28 completely bricks the app, the developer will not be there to give you an updated one (even if you're willing to pay), since most of the addressible market already bought the app in 2025, and after 2 years with almost no revenue, the developer stopped working on it, deleted the repo and moved onto another project. The developer can't even rely on a future OS update "encouraging" people to buy "App (N+1)" since it might be "ain't broke" for 1 year, or for 5.
The point of a subscription is not to rip you off, it's to acknowledge a few realities:
1. For reasons beyond developers' control, platform vendors do not provide a "permanent" platform, but a shifting one without any long-term guarantees. You can put a 100-year certificate into your app, but the OS vendor might decide that only certs with expiration less than 45 days are okay and your app no longer works unless you're around to (A) keep abreast of the platform's rules and changes, and (B) ship an update.
2. Many software offerings need to provide a server-side component, which is never a one-time cost.
3. Relying on upgrade purchases to sustain a product gives developers perverse incentives to shove a ton of new features just to be able to pitch "Upgrade to Appitron 2!" with a ton of bullet points, whereas subscription pricing incentivizes them simply to keep users loving the app forever, including adopting new technologies but also just improving the core experience.
Due to 1 and 2, it makes sense to let users who stop using the program after a short time pay very little, and to let users who rely on the continued operation of the program, pay a little bit each year, instead of paying $500 once and using it for a few years, and maybe upgrading for $250.
Reality is that many modern software developments have plenty in common with designing a toilet. You spend time identifying the problem statement, how you can differentiate yourself, prototype it, work out the bugs, ship the final product, and let sales teams move the product. The difference is the toilet can't be turned into a SaaS (yet) and, if it ever could, that would break functionality because you're supposed to poop in it, not have it poop on you.
Mercedes restricts the performance of some cars if you don't pay $1200 a year for the “Acceleration Increase”. You have to pay more if you want to use the power you already paid for.
BMW offer heated seats for £10 a month. The car has heated seats that work fine, and you paid for the hardware already, but they are turned off if you don't pay more.
Neither of these are anything to do with ongoing costs to the company, like support or mobile connection, they just want ongoing revenue.
If I have "Ajax" brand leather shoes sown by an East Asian sweatshop worker, who is the "creator" of the shoes, for purposes of benefiting from this system?
We are agreed that the company "Ajax" is not a creator, yes? Companies don't create - people create. Patented inventions are created by people, though patent ownership may be transferred to companies.
So does the monthly fee go to the skilled laborer who sewed the pieces together to give the final form? And also the laborers who turned cow hide into leather? As well as everyone involved in the shoe design? Does it also pass to their inheritors? For how long?
The house I owned was built in the 1950s by a local construction firm which is still around. There were several owners before me, including ones who remodeled and renovated it. Do all of them get part of my monthly fee? Or does it go to the woodworkers and plumbers and other builders who did the actual work?
I have books in my personal collection from authors who died decades ago. How do I reward Robert Heinlein in this "keep paying" scheme? Some of these books I bought used, so neither Heinlein nor his estate ever got a penny from me.
But that's fine, as the price point for the original sale already factored in the effect of the First Sale Doctrine.
Just like how the price of a car, house, bike, shows, etc. already factors in the reward for everyone involved, without needed an entirely new system to determine who the "creators" are, and how they get paid monthly.
And that's all assuming the fee distribution system itself is fair. We need only look to academic publishing to see unfair things can be once a system is entrenched.
The issue is a mismatch of incentives - customers wanting things for free - even if they aren’t actually customers. Vs businesses need/want for ongoing revenue (ideally for free too!).
Both sides are never going to be perfectly happy, but there are reasonable compromises. There are also extractive abusive psychos, of course.
Free customers can store 3 hours of sound. This former paying customer had more than 3 hours of sound stored.
The comment said SoundCloud was a terrible company holding their data hostage, by not letting them do anything with it except delete things to get it under 3 hours, and threatening to delete all of it if they didn't.
This part is left out in modern software development.
Bugs ? What bugs ? We just (re)wrote a new version. This one should be better.