Hacker News new | past | comments | ask | show | jobs | submit login
Subscription or no subscription? That is not the question (ia.net)
48 points by OberstKrueger 9 months ago | hide | past | favorite | 47 comments

This seems like a long winded mess of anecdotal opinions and assumptions, that still somehow ignores the real reason for subscriptions. The list of acceptable uses for subscriptions is silly, subscriptions are always acceptable if your customers accept them. And the authors refusal to try subscriptions out of a fear that short term revenues will decline is an abandonment of long term thinking.

And I am not a subscription apologist either. I just wrote a long post on the need for Apple to build tools for selling upgrades as good as the ones for selling subscriptions.

The real driving force for subscriptions is the need for recurring revenue to support a software business, especially bug fixes and enhancements. The one time sales model never worked in software, it was a path to oblivion that was quickly abandoned or never used by the vast majority of PC developers.

That’s why the PC software market became dominated by maintenance plans and paid upgrades. If you had a good product, and offered valuable new features in an upgrade, your users would be more than happy to buy the upgrade. That matched revenues better to costs, allowing software companies to keep more employees on staff to actually improve the product.

When Apple came out with the iOS App Store, they broke the upgrade model. Existing users get every update for free. Releasing a large update as a separate product means it can’t access the existing user data because of sandboxing, and it can’t be priced at a discount for existing users.

There are workarounds to all these problems using in-app purchase and data export/import schemes but they are costly in developer time and offer a poorer user experience than traditional upgrades did.

I’m convinced if Apple had an upgrade purchasing system as easy to implement as subscriptions, developers would abandon most of their subscriptions and replace them with upgrades.

I'm pretty surprised Apple has let it go this long. Without upgrades, everyone who made software for OS X before the App Store has been feeling the pinch.

It's hard to make money off of features you've improved substantially but most of your users bought four years ago. Except games, where you just change the name and a bit of the story and sell it again.

The part about separate upgrades not being able to access existing data is not accurate.

You can use a container that is shared between more than one app - this is not an import/export scheme.

This is true, but it’s not simple and requires quite a bit of work for both versions.

Microsoft offers Office as both a 1-time purchase and on a subscription basis. The price of the 1-time purchase is generally about what a subscription would cost over 3 years.

Since Office is the kind of tool you would keep for 3+ years and has an established reputation, this doesn't seem like a bad deal. Especially since Office has always been expensive.

People get angry about subscriptions when the subscription price far exceeds the previous 1-time purchase cost within a reasonable amount of time. In the case of the Android tool in this article, that's either $30 up front or $5 per year. So it seems that the subscription is either underpriced or the up-front price is too high: it's not surprising that they didn't have a lot of anger directed at them. Flip that around and say $15 per year or $30 up front, and I'd bet you have a lot of anger directed at them.

Note that MS Office has a cycle length of 3 years, while their cycle length is somewhere between 7 and 10 years.

How long do you get updates for the 'single year' edition of MS Office now? Does Office 2017 still get security/bugfix updates?

I forget what they promise to support, but MS are still pushing secuirty updates for Office 2010 - https://support.microsoft.com/en-gb/help/4563408/august-2020...

EDIT: Dug out the end dates for Office.

Ver - Mainstream Support - Extended Support

Office 2010 - no longer supported - October 13, 2020

Office 2013 - no longer supported - April 11, 2023

Office 2016 - October 13 2020 - October 14, 2025

Office 2019 - October 10 2023 October 14, 2025

As a small developer who lives entirely from software sales, a large part of which are in the form of monthly subscriptions, I think this author misses an important reason for using subscriptions:

    N. You want to smooth out the monthly income associated with sales, avoiding big peaks when you do major releases and avoiding large troughs when there are no releases for some time.
Put differently, if you want me to keep wprking full time on the software, help offer me some assurance that I'll still be able to afford to do so in a month, or 3 months, or 6 months time.

I just wonder what kind of software needs a full time dev once version 2.0 has been released?

What I mean by v 2.0 is major bugs have been found and major missing features have been added.

It seems to me that after that it's just the feature treadmill to justify asking money recurrently.

As a dev myself I understand you want to live of your craft, but as a customer I'm not going to pay all my life for an app, I will find an alternative.

DAWs (digital audio workstations) have no closed lifetime or defined feature list. The workflows and desired features expand every year. Every once in a while, someone comes up with a game changing idea that rewrites everyone's expectations of what is supposed to be possible (e.g. Ableton Live and their automated fitting of audio to the music grid).

If you can't convince your users to fund the development of your software with a subscription then perhaps it's not delivering real value to them. After all, if your users are indifferent to your software disappearing into thin air because you can no longer afford to pay the bills, perhaps they never cared that much about it in the first place.

I don't think this is true in the slightest.

I happily plunk down for a one-time purchase of software, because it's a simple question: is this worth $n bucks?

A lot of it just sits there for months on end, and then I have the task which I bought it for again, and I use it.

But a subscription? Now I have to ask if it's worth $n bucks a month on an ongoing basis. It's a whole different question, and I feel like if I'm not going to use it for a few months, I'm just bleeding money I'd rather keep in my wallet.

I'm especially wary of products which want $n a month to lock my creative output up in a proprietary format, such that I can only retain access to it if I keep shelling out. That... sucks, and it baffles me why it's so popular.

When I see a subscription I always ask: what's five years of this going to cost me? Why five? I dunno, that's just the multiple that makes sense to me.

Very few subscriptions on offer make sense on that basis. I'd have dropped Netflix a year or two ago, but my Mom likes it. I keep almost dropping Apple Arcade, a piddly five bucks a month, but there's this one game I want to finish and never get around to it... and there it is: a lot of subscriptions are sticky, in a bad way. As far as I know, there's no way for me to pay full price for that one game, and keep my save point, and ditch the rest of the subscription.

Now if I'm making money with it, that's a whole different ballgame. Sure, sign me up. But if I'm not? Likely as not, it's gonna be a pass.

> and it baffles me why it's so popular.

The Adobe situation is simply because it's such a good product suite. You can transition away but the pain points of using other software can be high enough to warrant paying for the software[0]. As for why the majority of users and businesses still pay for creative cloud, it's probably because they do see the value of receiving updates and bugfixes effectively forever.

0: https://youtu.be/L9VysWRHPdI

Adobe wasn't what I had in mind.

That's a moneymaker, for one thing. For another, they're the industry standard, and yeah the formats are proprietary but if you take a couple years break and start paying again, your files are going to open.

I was thinking more of products like Roam Research. Not to pick on them in particular, I know a lot of people like it, but it's just a non-starter for me.

The Moleskin iPad app is the one I'd most like to use, if they came up with a different business plan for it.

One time fees make software development into something resembling a pyramid scheme. Software development costs are recurring, if you hire someone they get paid every month not just once. So the developers must continually find new customers in order to support the old ones. At the point when they can no longer do so, the scheme collapses and everybody loses out.

I understand that, but we had a solution to this which worked fine: upgrades.

A number of trends have made this unpopular, and yeah, Apple deserves a big chunk of the blame, but by no means all of it.

But this has been explored fairly well elsewhere, so I'll refrain except to say: I do hope the steps are taken to reverse this unfortunate trend, because shoehorning subscriptions in where it's a bad fit is no kind of solution.

I like Jetbrains' subscription model. Haven't ever not been on the subscription. It just gives me peace of mind knowing I won't be left out to dry if I need to be off it.

For those who are not familiar with JetBrains' subscription model:

> A perpetual fallback license is a license that allows you to use a specific version of software without an active subscription for it. [0]

There are some details about which version you are eligible for. The spirit is that if you subscribe for ≥ 1 year, you have some version available when you stop paying.

[0]: https://sales.jetbrains.com/hc/en-gb/articles/207240845-What...

JetBrains was initially going with a more traditional subscription model, and introduced the perpetual license fallback in response to negative feedback.

HN comments and the original announcement: https://news.ycombinator.com/item?id=10165334

The followup introducing the perpetual fallback license: https://news.ycombinator.com/item?id=10278285

Yeah I was a bit unhappy but accepted it for what it was when it was announced. I'm glad other people complained. I really like the current form.

That’s actually pretty smart, but if only 1 year of subscription is equivalent to a license, it feels like the subscription is overpriced (but I know nothing about the prior price)

Their software is really good and every year gets you more so I've never felt like my $150 has been poorly spent. I could use the old version, but honestly I'd much rather use the newest.

Thanks. I was just going to ask what made it unique.

We use iA Writer daily at satchel.com. We pay a one-time cost of $30 per seat, but I think we get much more than $30 worth of value from iA Writer. So when I read the article, I thought that the $5 per month price is much too low for the value we get out of the product. And then I realized it's actually $5 per year.

Which brings up an interesting observation: the folks at iA seem to be tech purists at heart, and seem disinclined to ask for more money for their products. But I suspect that people would surely be willing to pay more money, and that if they were to price-segment their customers (e.g. by building out some collaboration features for teams, which we would love to have, and then charging a premium for that), they might be able to effect a step function in their business.

I always worry when a beloved app seems underpriced, and wonder how long they will be able to stay in business.

I don't think you representative of a lot of customers.

Where I work I must do without critical software because it's deemed too expensive by the management (weekly restaurant outings for the managers aren't deem expensive though ;)

I think that's okay though. If we're only representative of only 5% of the iA userbase, but we're willing to pay more than 20x of the subscription price ($5/yr → $8.33+/mo), then the 5% of users like us can double iA's revenue.

This is an interesting statement "Basically, Android was always a struggle, but for the first time since the very beginning, Android is moving towards sustainability."

And I didn't see any mention of the economics of iOS. From everything I've heard, iOS users are willing to spend more money. In that sense, it does seem like Apple is bringing something to the table with its 30%.

iOS devs make 50% more than Android devs, and the average iOS user spends about 10x more than the average Android user does on Apps.

It's strange how that works out.

I use both Android and iOS and I've easily spent 10x as much on iOS apps as Android apps.

Why didn't you spend that much on android apps? Lack of good apps that you would pay for or enough free apps?

I have done the same, here are some the reasons.

1. Better targeting by app companies when I am on iOS device.

2. Trust in the Apple app eco system

3. Most top apps are paid, so it makes it more acceptable to pay

That's why Apple is so afraid of sideloading. Because it threatens their business model.

And their developers business model, and their high customer satisfaction levels.

Unless it's for something fairly trivial, I generally don't use free apps.

When I do buy an app and have a choice I generally will go with the iOS version. I haven't given it a lot of thought, but I suppose it comes down to a couple of reasons.

1. I have never really trusted Android. Probably because it's made by an advertising company and as time goes by, the ad supported business model is starting to feel more and more unhealthy to me.

2. Android apps are rarely as polished as similar iOS apps. Even for relatively basic things like layout or color scheme, iOS apps seem to have more care put into them.

Not being familiar with these things, why not just emulate "upgrades" through in app purchases? You would offer a limited version of the app for free, and then prompt users to pay for the full version. Every now and then, you would declare a new version of the app, and only give access to new features you add to relatively recent users. Older users would be prompted to pay (again via iap) a reduced price to get the "upgrade".

Issues with this are:

- still has the 30% fee instead of 15%

- adds complexity, due to needing to track what each feature is, and which users are allowed to access each feature

- to obviate above, you would probably end up giving all bugfixes/optimizations/etc to old features to all users.

Could maybe clear up the first point by accepting payment via subscription instead, with the following caveat: "if you stay subscribed for more than 3 years [say], you will gain permanent access to the app, minus any features added after you terminate your subscription". As a user, I would feel pretty good about that (i think), since I'm not locked into either a subscription or a full payment. Of course, this is likely a downside for whoever makes the app, since it probably entails less revenue for them.

I like the pricing model of language apps like LingoDeer and Rosetta Stone. They offer a lifetime subscription for a one-time fee that costs less than two annual subscriptions. I have no problem taking that offer, but I have no recurring subscriptions for apps.

I love that model, I can subscribe for a month to see if I like something and then either go lifetime or dump it.

Our app (https://about.domestica.app) uses a yearly subscription model for both SaaS and self hosted users. For self hosted users, the licenses are permanent/perpetual, but tied to a minor version (i.e. the last release before your subscription runs out is 1.1.0, you can get 1.1.1, but not 1.2.0, etc).

We've played around with different pricing (monthly, tiers) and none of them changed conversions much. What did help was replacing the free trial (no CC required) with a free tier, as it provided more of an onboarding journey and the subscription purchase becomes the next logical step.

Prior to monthly or annual subscriptions, business software was purchased for a larger lump sum up front and perpetually owned. If you wanted to get updates and support you often paid a yearly maintenance fee ranging from 10% to 30% of the initial license/software cost. Saas, Software as a service was originally paying a monthly or annual fee for the software maker to also pay and run the servers. Now it seems Saas stands for software as a subscription where that still applies but everyone expects continuous never ending features to be added.

For corporate software maintenance as mandatory, making it little different from subscriptions.

I've been thinking about this quite a bit, and though the DropBox is mentioned briefly, I'd really like to understand more about why we're comfortable with paying a subscription to them, but not other similar apps.

We're building a wearable, so we're in the hardware space, but there is processing of data and personalized improvements to each users experience happening every day. I think we're similar to Whoop in this way. Any Whoop users want to chime in about how they feel about the subscription model?

If anyone else is finding the style hard to read here's a link to the outline version (you can do it to most articles by just prefixing the URL with outline.com/)


Hard to see how apps like this exist. The author seems to enjoy writing about his app as much as creating technology that exists already, and has for decades.

For $7/mo, you can have MS Office. You get a real Word Processor for all possible writing that happens in the real world where proper want to collaborate on business documents, or use Focus Mode for distraction-free creative writing. IA makes a big deal about their grammar checking but so what? Other word processors have had it for years and MS’s is genuinely impressive when you learn how to set it for your needs. Plus for the same price, you get the workhorse beast Excel, a business email and calendaring client (Outlook), the #1 presentation tool (PowerPoint), and a highly functional note-taking tool (OneNote).

I’m not trying to push Microsoft’s product line. There are others like Corel, or Apple’s iWork suite.

I just don’t understand the plethora of “distraction-free” editors, or all the fuss about code editors. Just use VS Code, JetBrain, Atom, or what have you. But as we speak, people are sweating to create yet more code editors which will have twice the bugs and half the functionality of the existing ones.

The world of writing tools is even worse. Markdown is a bit of a joke in the real world. If I see one more article about how someone wrote a book or dissertation using markdown, python, pandoc, makefile, and git... sigh! The reason they write those blog posts is because no one would believe anyone would go through the hassle of avoiding Microsoft Word to that extreme. It’s one thing to be a bit “retro” and keep using WordStar or WordPerfect, but those are at least real word processors.

A whole lot of people don't like writing in Microsoft Word, for a variety of reasons, and there are editors of varying types aimed at that audience. A lot of professional authors write in Scrivener. I know of more than you (apparently) think who write in Ulysses or iA Writer, which are both pretty capable editors. I own all three of those apps, am actively using Ulysses for a big fiction project currently, and don't own Microsoft Word. I know of more than a few fiction authors who like writing their prose in text editors -- Charlie Stross, Cory Doctorow, and Steven Brust come to mind off the top of my head. (Brust has used Emacs for a very long time and I presume still does; Stross and Doctorow both used BBEdit at some point in the past, and I believe both used Markdown, although that was years ago.)

As for whether Markdown is "a bit of a joke in the real world," I suppose it depends on what your definition of the real world is. Having attended a major technical writing conference a few weeks ago, I can assure you that while there are arguments in my field over Markdown, they are not over whether we should just give up on it and use Microsoft Word instead.

They're talking specifically about an Android app. Almost none of the apps you referred to are available on Android, let alone with the features you mentioned.

As for the Markdown vs Word doc argument, sometimes you want to do things with your data that just isn't easy or convenient if it's inside a binary/proprietary filetype like doc/docx. Maybe that's not a use-case that you've ever dealt with, maybe it's a "doomsday" scenario that you think is unrealistic and over-reactionary, but really who cares? There's no need to "yuck someone's yum", just because you don't get it.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact