The Economics of Productivity Software

I am archiving older pieces I have written on other sites, making this the definitive home for all my work. Enjoy!

Since being terminated in May, I have founded my own independent software vendor business and begun work on a productivity application targeted at artists, and animators in particular. This is an idea I’ve been working on for years, and I’ve devoted a nontrivial amount of time to considering my go-to-market strategy, among other concerns. For a variety of reasons, I have decided to lead development off on the Mac, with versions for other operating systems (including mobile) to potentially come later. As a consequence, I’ve looked very closely at the pros and cons of the Mac App Store, and have a complex set of feelings about it.

One of the longstanding developer complaints about the Mac App Store is its lack of upgrade pricing. (This complaint is also made about the iOS App Store, but I’m personally less interested both because I’m not making a mobile app at the moment and because the effective price ceiling for “modestly-priced” apps is a mere $10. My product, for the Mac, will eventually be priced somewhere between $50 and $100.) This has generated a fair amount of discussion, including a segment on the Arment-Liss-Siracusa Accidental Tech Podcast, and blog posts by "Underscore" David Smith, Elia Freedman, Justin Williams of Second Gear, and, just today, Steve Streza of Pocket fame.

Freedman makes a compelling argument that the issue isn’t price, but the complete lack of product differentiation. His argument is that offering trials would allow users to perceive product differences through use, justifying sustainable prices for product. That’s an important part of securing that first sale, but a paid upgrade captures additional revenue from an existing customer in exchange for additional functionality/value, and discount pricing on the upgrade boosts the conversion rate on that proposition. Why does this matter? Because the value that a user gains from the first purchase of an app dwarfs the value gained on subsequent purchases, discounted or otherwise.

Let me repeat that: going from not having to app to having any version of the app is almost invariably more valuable than going from having any version of the app to any subsequent version of the app. To argue that you can charge the same price for vastly different levels of value is to disrespect the customer, unless the price for the app falls below a threshold of consideration.

Apple has powerful incentives to commoditize and devalue software, while tying that software to its platforms. Apps make its iPhones and iPads, and to a lesser extent Macs, more attractive as consumer electronics appliances. Having lots of cheap or free apps is great for Apple, but very difficult for all but the most riotously successful (in terms of sales volume) developers to sustain.

Remember when Adobe announced that Creative Suite would end with CS 6, and subsequent releases of its media editing applications would be exclusively available as part of its Creative Cloud subscription suite? Severe backlash, including a petition insisting Adobe backtrack, signed by over 5,000 “customers” (no way to actually verify this). Yeah, right.

Software-as-a-Service (SaaS) has been coming for a long time. Enterprise workers from the 90s will remember organization-wide attempts to deploy software on demand, like Citrix and aspects of Windows NT. We got Platform- and Infrastructure-as-a-Service first, in the form of Amazon Web Services, Microsoft Azure, etc. Well, the network is robust enough to support data, authorization and commerce transactions that connect a local client install to a remote server in a tethered mode where the client’s operation can be dependent on a paid-up account. The problem is that SaaS resolves developer/publisher concerns—smoother recurring revenue, greatly reduces piracy—but doesn’t provide any specific benefit to the customer.

We’re in a situation where developers are responding to the platform owner’s attempt to commoditize them by changing the purchase process to rent-seeking, and offering the customer nothing but rhetoric in return. Some developers try to add server-based features to justify the change, like Adobe’s inclusion of TypeKit’s full portfolio of fonts with your Creative Cloud subscription, but the customer owns nothing anymore.

Steve Streza’s post, in particular, bothered me and spurred this response. He assumes bad faith on the part of developers interested in discount upgrade pricing, characterizing their relationship with their customers as exploitative and abusive. I find this to be a common pattern among developers who have primarily built “social” and web-based applications, as opposed to local applications focused on productivity uses.

If your app isn’t compelling or crucial enough to be priced beyond the threshold of consideration, if you’re only charging $5 (or $15 on the desktop), then you can charge full price for each version without seeing significant audience attrition or customer backlash (beyond those assholes who complain when an app costs $1.99 instead of 99¢).

At the $60, $70 level, when your name isn’t Autodesk or Adobe or Corel? At the $200 level, like OmniGraffle Pro? You need every sales trick at your disposal to reward your loyal customers while showing new customers how you will reward them in the future. Apps priced at this level are because they help users create work, often salable work, that they can in turn profit from. The value to their users justifies the price level, but the price point makes users cautious about that much investment.

When upgrade pricing comes under attack, and when App Store operators insist on a sustaining share (30%!) of any subscriptions sold through their interfaces, developers adapt by turning what were local install-only applications into client interfaces to server-side components and requiring an existing sub. Adobe’s Creative Cloud apps can come to iOS or the Mac App Store in full now, because you’d need to have active Adobe subscriptions (purchased direct, much like Netflix) to use them.

Upgrade pricing isn’t exploitative or irrational or a cheat. Developers clamoring for it in the Mac and iOS App Stores aren’t holding on to an archaic business model, they’re making the mistake of thinking Apple cares about their success.

As for me, I plan to offer my product in the Mac App Store, but perhaps only a “lite” version. I’m hedging my bets, and also exploring web-based services I can launch and tie my product to, generating subscription revenue to boost downloadable sales.


Since this article was written my position has "evolved" somewhat. I now believe that discrete user-facing versions are anachronistic and irrelevant to users, and developers need to find sale/delivery mechanisms that completely hide them from view. Specific to the iOS and Mac App Stores, I am exploring the potential of leveraging In-App Purchases to sell and deliver individual and bundled features, in effect offering upgrade paths to "version N" through a rolling series of add-ons to its predecessor.

This allows users to pay for only those features they want, and provides me with feedback on what those features are. And if sweeping architectural changes are required to support new work, I can remove older IAP/bundles from sale and bundle in their equivalents to the main app's (free) software udate, restoring a tabula rasa for future enhancements.