A Critical Analysis of the Subscription Economy
medium.com
medium.com
This has allowed smaller software applications such as Sketch, The Affinity Suite and Capture One to grab marketshare. I think Sketch’s current business model is perfect. Customers pay for a year of updates. If a customer decides not to upgrade, they stay at the same version as when their license expires.
I’m certain Adobe has made a good amount of money in the short term, but they've forfeited a monopoly. I’m more skeptical about their long-term business.
EDIT: Updated for clarity.
It seems you lose editing access, you still can view them. Still not great, but I wanted to clarify.
I suppose you could use a PSD or AI to Sketch converter. I'm unsure about other file types.
I am not arguing that subscription models forced by Adobe on users is a good thing. It is, IMO, horrible and is the reason I am still using CS3 for my photography hobby.
It was not pure fault of subscription model though. ReSharper became too slow in the recent years.
Photoshop Illustrator and inDesign haven't had upgrade worthy releases since like CS5. Acrobat Pro seems to be getting worse.
So you pay for a year of updates, but if anyone you collaborate with updates their software, you have to as well, because you can't open the files they produce or even edit and save.
Most of the time you're paying for updates so that you can open collaborator files, not because you want new features. It's effectively a subscription paid annually.
I don't think they ever would have had a monopoly, even if they had kept their perpetual licensing instead of moving to subscription. The cost of those licenses was pretty high, so there was always room for someone to come in at a lower price point and undercut them.
Besides, $10/month for Photoshop and Lightroom is much more accessible than $699 for a perpetual license, especially if you don't need to use it that often.
I like to think those capabilities jumped because the new influx of Adobe refugees provided them with enough revenue to supplement their dev team. Or am I wrong?
I would argue that the more psychos in the market, the better for startups and I am quite comfortable with that idea.
At 5% interest, a subscription of $1 a month is worth a fixed cost of $240. Photoshop and Netflix's $10 a month is worth a fixed cost of $2400. Xbox Live's $60 a year subscription makes it worth about $1200, making it cost 3 times more than the console itself. World of Warcraft's $15 a month subscription comes out to a whopping $3600, making it 60 times more expensive than a normal new game (if you wanted to subscribe to it forever.)
Or put another way -- what?
https://en.wikipedia.org/wiki/Time_value_of_money#Example_2:...
But their calculation was ... let's say wrong :-)
Seems reasonable, but I'd also stop using most subscriptions before 20 years: 5 years seems more likely. I'd probably cancel a subscription if I stopped using the product after a few years or if the related hardware died.
Well, yes and no. GP is calculating the discounted value of receiving $N annually forever (assuming a fixed constant interest rate r):
N/(1+r) + N/(1+r)^2 + N/(1+r)^3 + ...
= N/r
(You can derive it from (for |q| < 1), 1 + q + q^2 + q^3 + ... = 1/(1-q) )
Note, by the way, that currently rates are very low (say 1%), and thus receiving a perpetual stream of $N per year is very valuable, namely 100N. That makes the switch to subscriptions even more valuable to these psycho firms.
As you say, they're lending you what would've been the upfront payment for the software. They receive your subscription, and have to pay (in interest, to their bank) very low rates. So, on the (fixed) notional they currently pay low rates, while receiving high-ish subscription.
In an ideal market, you would expect the discounted subscription to match the upfront payment, which would mean that the "high value" is simply the price that they have to pay to get those future subscription payments (or that they could sell those future subscription payments for on some sort of "customer subcriptions market" if there were one). Once they've bought them, they also have the risk of rising interest rates wiping out the value of that investment.
(And also, this doesn't really apply all that well in this case anyway, because your typical software license subscription is not a perpetuity, the span of time that people will want to pay for using the software as it is now is rather limited, and they can pretty much stop paying at any point, while you'd supposedly have spent the loan from your bank on your programmers, and your bank will still want to get their interest payments.)
I think this is something people don't often think about. But it makes it very easy to compare different costs and benefits that involve time. E.g. imagine a building is costing $1000 a year to maintain, and a new building would cost $50,000 to build. You can figure that it would be easier just to invest $20,000 and use the interest to pay for the maintenance. And save $30,000.
But if building a new building cost only $10,000. It would be a good investment. And you would expect to save more money than you would gain from anything else you could invest in with that money.
The same logic applies to subscription fees. If a software license costs either $120 a year, or a fixed price of $5,000. You should prefer the subscription. Hypothetically you could just invest $2400 and use the interest to pay the subscription. Saving $2,600.
Except you do not actually use software forever, and there is also no market for licenses for 10-year-old versions, so you have to actually calculate it as an annuity.
So, if you expect to use the software for ten years, the NPV of 120$ per year would be about 900$ at 5% interest.
Also, it makes much more sense for a large corporation to be exposed to the risks of investment return, i.e., it makes more sense for fixed-price software licenses. Well, maybe. Whoever has a smaller fraction of total expenses over cash reserves, and is this able to ride out a recession, should get the risk. In the case of a person with a big retirement fund, they should get the risk. However, in the case of someone living paycheck to paycheck, the company should get the risk.
However, coming up with some number X prices for various cohorts is not particularly useful, since simply introducing choice confuses some users and causes drop-off, let alone optimizing those choices and explaining them.
1. Users expect their apps to work on their computers, tablets, and phones, which generally means you need to provide some ongoing service to keep your data synced between all of the devices. (For some simple apps you can get away with using a third-party service, e.g. Dropbox or iCloud, but many apps with complex data will need their own custom-built solutions.) Depending on the kind of data you are storing, this can get expensive.
2. Users expect perpetual upgrades for every new version of iOS/macOS/Android/Windows that comes out. This obviously requires development work, especially since iOS/macOS/Android are annual releases at this point.
These ongoing costs imply that some kind of recurring revenue is necessary to keep the software running and up to date, which is why the move to subscription revenue has become more popular lately.
Besides, I question the premise that "perpetual" licenses have ever truly been perpetual except in name only: eventually that software you bought will stop working on newer operating systems and devices and you'll have to buy a new version anyway.
2. https://www.youtube.com/watch?v=sxXs0Yy5-0Y For 2, yes, users may get mad when some vulnerability ends up leading to a leak of PII or worse, but to claim they "expect" it, again, no.
2. I'm not talking about OS updates. I'm saying consumers expect their apps to be updated to keep them compatible with new versions of their OS. That means someone that bought your app for iOS 6 five years ago absolutely expects it to be updated to work on their new iPhone 7 running iOS 10.
Besides, Windows is probably the outlier here. People might hate Windows 10's automatic update system (which was clearly implemented poorly) but a lot of people update to the latest iOS, macOS, and Android versions right after they're released.
I don't like the subscription model because in the end, for multiple subscriptions, it all adds up and ends up being more expensive. And that after spending a lot of money on a subscription you still don't own the software.
(Not to mention "season passes" for games.)
Sure, they provide revenue that can support security fixes - but at the cost of decimating the incentive to improve features to generate upgrade revenue.
After all, if spending more on development doesn't raise revenues much, but cutting down to a skeleton team of maintenance developers means that subscription revenue is almost pure profit, there'll always be investors pushing for that.
Seems to me that invites our industry to stagnate and stop producing value, if you don't need to make a better product in order to bring in revenue.
I like the subscription model, especially if it has a free tier, because passionate customers can support the developer team and subsidize less passionate users
But you do realize that the subscription model is exactly the thing that removes the incentive to do that, right?
With the subscription model, the company is still going to make that income regardless if they have a new product. As long as no other competitor has developed a competing product, there's no reason to innovate. (and there's lots of ways to deal with competitors besides innovating)
I'm not particularly happy about moving to what is effectively a rent society, but I think this is one of the few silver linings of the subscription model.
It's almost impossible to keep mature products going without a subscription model. Mature products eventually only need:
a) support and b) compatibility/stability/performance updates.
But, these types of updates can be very expensive to develop and are very difficult to market as a new release (works with XXXX now isn't a very compelling proposition for upgrades).
It's kind of the same problem that open source software faces after it matures. Nobody wants to just sit and do support/incremental changes on old software, at least not for free.
From the development side of the business, the more frequent release cycle has allowed the product teams to be much more responsive. Under the old model, with major updates every 18-24 months, I might receive a feature request which is implemented in a few weeks, but which the user doesn't have access to for a year or more. This was hugely frustrating as we want to make features and products our customers want to use, and to have it ready and waiting for so long was not a great experience for anyone. Now, we can have that feature in users hands in weeks or months.
I don't understand this idea that there will be reduced incentive to release updates under a subscription or membership model. If anything, frequency and responsiveness feels more important than before, and I don't know of anyone here who feels differently.
edit: I was more on the fence about subscriptions than unsupportive. It was a drastic change that required either gradual transition or a rip-off-the-bandaid approach.
Buying a license today is investing in future development. I wouldn't pay to use software that's not actively being maintained. Nor would I expect my customers to either.
What is discussed here is the case where you lose the right to use the software if you don't continue to pay, completely independently from what that money is used for, which is why it removes one important incentive to improve the software.
And if your skillset and your data is locked up in some software, it's questionable whether you really would stop paying, and thus stop using the software that you depend on.
"...only reason they'll buy a new copy is if the new one is substantially better than the old..."
The obvious, intuitive cause & effect isn't always the case.
I bought software to encourage future development, vs wait for new release then decide to upgrade. Push vs pull.
2. That people are willing to effectively pay in advance for a future release without any guarantees what they will get for it is irrelevant to the question of whether you have the lever to shut off that cash flow if they continue to not deliver any updates of value to you, while still using the existing version.
In other words: You are simply missing the point.
What does change is there isn't the incentive to pile in dubious features to justify charging for an upgrade every other year.
Can you name a single case where this has happened with subscription software, though? The big named examples in the article (as far as I could stand to read it, at least) are Adobe Creative Cloud and Microsoft Office 365.
Here's the What's New page for Photoshop CC 2017, along with links to the changes in various bugfix releases that have occurred since release:
https://helpx.adobe.com/photoshop/using/whats-new.html
You can find the release notes for the past few releases of Photoshop:
https://helpx.adobe.com/photoshop/photoshop-releasenotes.htm...
They aren't sitting on their laurels. And Office hasn't stagnated, either. If the subscription model removes the incentive to do that, why are they still doing that? You get new features in Office 365 sooner than you do on shrinkwrap Office releases. Adobe products have been subscription only for over four years now. The incentive seems to be in place.
There are other examples as well. Do you own Visio or Project and use Citrix? You can't use it unless you convert to a subscription.
Most of the office apps changes have been around supporting other office components. Stuff like federated login and not crashing. :)
> Can you name a single case where this has happened with subscription software, though?
Every single one of them. It's just a matter of logic that that incentive is removed. There might be other incentives that still keep them innovating or whatever (like, say, competition, if it's reasonably possible to migrate away), but that one just logically isn't.
Yes, of course, essentially nothing is monocausal, but that doesn't mean my point was fundamentally wrong: The subscription model (and by that I don't mean the setup where you have a contract for regular updates that you could just cancel and keep using the version you have now) removes the old version you already have as a competitor to the new version that you could pay for. That is most likely not going to be to the advantage of the customer, whichever form that disadvantage may take.
With the model where you can just keep using the old version without any further payments, you as the customer have at least two options two choose from: keep using the old version (doesn't cost anything), or buy the new version (you have to pay). Those two options effectively compete for marketshare, and as such, the latter option has to convince you with something that you consider valuable enough for you to pay the higher price. There might be further options in the marketplace, but that's not guaranteed, and also, migrating to a different solution usually has costs associated with it (learning the new software, lower productivity in the beginning, downtime, paying for the export/conversion/import of the data, ...), while the migration costs between versions of the same software tend to minimal, in particular lower than migrating to a different product altogether.
Now, with the subcription model, the first option is removed from the market. You only can buy the new version, or pay the migration costs of switching to a new product. Which is also not lost on the company selling you the software: If their market research people are any good, they'll set prices such that it's a bit cheaper than the competition plus migration costs, effectively turning a large part of your potential migration costs into their profit.
A real update would be better live editing. When they launched lightroom I was hopefull some of that workflow would make it to Photoshop. Nope. When they bought speedgrade I thought "amazing, I hope they make a print version." Nope.
The same is true for Illustrator, inDesign and Acrobat Pro.
That doesn't require a subscription. Plenty of companies offer a permanent purchase that comes with a year of updates or all minor-version updates or similar.
The option is there to treat it similarly to a subscription, but importantly you never have to pay and what you already own can never be changed or taken away.
- A year of support on our forums (real company-provided support, not just peer-provided).
- All minor releases and bug-fix builds within the year.
- After the subscription has expired, you still keep everything that you purchased forever, including being able to download anything you own at any time (we keep track of all of that for you).
We do have some "catches" to prevent free-riders at the expense of customers that maintain subscriptions, notably that you can only renew a subscription at the normal price within 2 years of your last renewal. After that, we start to ratchet up the renewal costs until, at the 5-year mark, you'll need to (effectively) purchase a new license.
Also, we offer inexpensive direct phone/email support through a separate "support session" method that is not tied to your subscription. So, even after a subscription expires, you still have direct support options.
The key is that any arrangement be fair and beneficial to both parties. In our case, we get a steady, predictable stream of revenue, and the customers get support, enhancements, and bug fixes quicker and at less cost.
This, for the vendor, is the other edge of the subscription model. If another flavor of the day arises that costs less, your customer base can swiftly move to this new contestant, especially in the enterprise space, because unlike a fixed license, a subscription isn't a capitalized cost. For Adobe, with their massive ecosystem, this is less practical for their customers, but a password manager? There are already a host of commercial alternatives that to an end user are all more or less the same.
It's true that we pay more for stuff now vs. perpetual licenses (it's prob a book's worth of material, but "opt-out is better than opt-in" and Houshalter's point on chunking are most of it). But in exchange, companies have the ability to invest that greater profit to a greater degree in innovation, so in the long-term we, customers, should get more out of it than we lost in dollars.
First an observation: Adobe's decision to increase profit secularly (i.e. beyond any accounting nuances at the time of switch) by switching to subscription is actually TWO decisions. 1) the switch in revenue mechanism itself and 2) the decision to NOT invest the additional profit.
#2 is the crux of my point, but the "to a greater degree" thing is a 1+1=3 effect:
"more profit" is straightforward because subscriptions increase dollars in, often more than making up for any churn. That cash could be invested to accelerate innovation. It's 1-to-1 apart from timing issues (queuing "yearly up fronts" :D)
"to a greater degree" - this is the special part. It is easier to spend more when predictability is higher. Or put another way, volatility is the enemy of spending to the "hilt". So companies like Netflix can spend (for our sake) more to the hilt when you have a tighter distribution as to where the hilt could be (vs. perpetual products which have poor visibility). Just ask any marketing person or sales person about how easy it is to spend in channels when attribution isn't a problem. BTW I define hilt as "as much as necessary to stay ahead of threats of competition/churn"
But yes, the company does have to choose to do this. Some are more relentless for their customers :). I just heard about the www.relentless.com thing lolz...
Quick aside: I saw this sick chart that compared 12-month lifetime value between Dollar Shave Club (subscription) and Harry's (previously per unit). Even though Harry's has always been branded as a more premium product, DSC made more money by the 12th month (even after churn). It's easy to forget that monthly click -_-
This then is the fundamental problem, when you finish the software do you fire the engineers that worked on it? If not, what do they do?
The fact is that software can actually be "done" and it happens every where. And it takes a special kind of engineer who will work on simply removing blemishes (bugs) and not write any new features.
And that fundamentally points at a really really deep problem, engineering software rather than just writing it.
If your software has very well specified and defined interfaces, and no 'escapes' then you can say it is engineered to be partitionable. And an engineer can be tasked with fixing one part, they can fix that part, and do the qualification tests for that part and then when they integrate it they can be assured the system as a whole continues to work as it always did, but now with a slightly different part. Not surprisingly this was a part of, if not the major part of, the notions in object oriented programming. Once the part (object) worked, you didn't have to think about it any more. An auto-mechanic doesn't have to know the physics of a an ignition system to replace a spark plug, a spark plug manufacturer doesn't have to know the geometry of every cylinder to make a spark plug.
Software as books. You don't go to the library and periodically buy updates to the books you own. Textbooks pull that off but even they know its really a scam. Pythagorean's principle hasn't actually changed and the last text book was just fine.
Open source kind of wins and kind of loses here. The poster child is gcc. Looking at the code base changes over time for a given architecture they start with a flurry of functionality changes, then smooth out to just bug fixes and the occasional optimizer tweak, and then when the noise floor of change frequency gets to a certain level you start seeing really sort of adhoc changes or perhaps changes that only help a very very small part of the user base. Its hard to capture both a gcc for ARM9 and gcc for ARM9 embedded in an FPGA fabric that can have optimized bespoke instructions. And yet they will try.
The author calls them psychopaths but they are the current generation of entrepreneurs which are building businesses based on rents rather than sales. Its all the rage, is not Uber a rent seeking business that simply takes a tax or rent for you using their app? The 'sale' was the car ride, the 'tax' is everything else. This works because collecting rent of $1 a month from a million people is "easy" they hardly miss it, and you're $12M/year richer for it. Selling them $12M worth of value is so much harder. Why push yourself? Everybody is doing it, the streaming company, the phone company, and even the maker of your coffee machine. If you haven't got an ARR you aren't a 21st century company these days.
It's one thing to rail about the varios psychotics or psychopathics of this that or another. That's not the same as understanding the dynamics through which a situation emerges. What's particularly disturbing is when it appears that the outcome is inevitable.
I.F. Stone, writing in 1967 on the Arab-Israeli crisis:
The essence of tragedy is a struggle of right against right. Its catharsis is the cleansing pity of seeing how good men do evil despite themselves out of unavoidable circumstance and irresistible compulsion. When evil men do evil, their deeds belong to the realm of pathology. But when good men do evil, we confront the essence of human tragedy.
http://www.nybooks.com/articles/1967/08/03/holy-war/
Let's look at the problem here.
In a commercial software world, there's a mismatch between cash flows and development. Worse, there's also a conflict between market mechanims based on marginal-cost pricing, and the long-run average costs of development. There's also the tremendous variance in customers' ability to pay -- price discrimination -- particularly for enterprise software.
If you sell shrinkwrap, or some other form of buy-once software, then sustaining the development efforts for the next version is ... difficult.
The two largest consumer softare companies of the 20th century, Microsoft and Apple, both sponsored that development through hardware sales. Apple did so directly, by selling its own hardware. Microsoft did it indirectly by way of per-CPU licensing of IBM-compatible PCs. Both companies avoided the significant costs of direct software sales.
The concept of recurring-subscription revenue is usually associated with periodicals, though that is a relatively modern development. The term doesn't emerge until the 19th century (previous usage was generally in a religious context), and it generally referred to stock subscriptions. Another variant was the subscription library.
(See links below.)
In a magazine subscription, you pay for the right to receive fresh material, but continue to possess any previously received issues.
The model for software subscriptions was in large part IBM's practice of leasing rather than selling computer hardware. Phone systems often followed similar practices. Hardware and software occupy different worlds in that hardware is fairly intrinsically limited: you have a computer, or perhaps a rack, or aisle, or datacentre. But these are unitised, and you're not individually leasing, say, hard drives, CPUs, memory cards, or capacitors, within the computers.
My Debian systems typically have a few thousand individual software packages installed. For a proprietary OS, that number falls, but is still considerable.
Dealing with individual software packages on a subscription basis from here to eternity is itself a major complexity problem I'd, frankly, rather not have to deal with. It may work in instances, but not at scale.
At the same time, there are the financing and cash-flow problems of software developers.
How do you bridge those divides?
https://books.google.com/ngrams/graph?content=*_NOUN%20subsc...
https://books.google.com/ngrams/graph?content=subscription+*...
William R. Catton, Jr., has a concept he calls "futile vilification", short version: "There is no point to another morbid wringing of hands over mankind's alleged "greed" or immoral myopia."
Longer, quoted here: https://www.reddit.com/r/dredmorbius/comments/2v3251/william...
Donella Meadows has a concept of 12 leverage points. "Changing personnel" doesn't even rate mention.
https://en.m.wikipedia.org/wiki/Twelve_leverage_points
The history of IBM, AT&T, and Xerox focusing on leasing rather than selling hardware is useful, as is a reading of the history of the publishing industry. Software is, in some ways, a development of that.
Thank you for this, I do intend to drill down into the subject a little more and this will make for good reading, as will this entire thread.