The Mental Blog

Software with Intellect

1 note

Why is the Mac App Store really failing?

At various times since its introduction three and half years ago, discussion has arisen over the viability of the Mac App Store (MAS). It’s flared up again in recent weeks, with a post by Milen Dzhumerov capturing the essence of the criticisms.

I have lots of respect for Milen — many aspects of my Ensembles sync framework are based on techniques used in the Clear app, which he developed — but in this case I disagree with his conclusions.

In fact, I am probably at odds with most developers on this, and have been many times in the past. Wil Shipley expressed very similar views to Milen in 2012. I replied in a similar tone at that time. A year later, I was on the iDeveloper podcast, again flying the flag for the unpopular cause, this time against Ken Case of The Omni Group.

There are a couple of ways I tend to be at odds with other developers when it comes to criticizing MAS. First, I believe the old shareware model of selling software, requiring customers to fill in forms with credit card numbers on obscure web sites, copying serial numbers, paying upgrade fees etc, is antiquated and should be deprecated ASAP. It’s a bad experience for the customer, and developers have an obligation to find ways to improve it.

I also don’t think that MAS needs to adopt these traditions to succeed. I don’t think it needs trials, upgrade pricing, non-sandboxed apps, developer responses to reviews, or any of the other solutions being proposed. I doubt any of these changes would have a significant positive effect on the Mac App Store. I doubt the problems with the Mac App Store are technical at all.

The Proposals

Before I get to what I think the real problem is with the Mac App Store, let’s try to debunk a few of the standard solutions being put forth.

Free Trials

Apple could easily have done this by now if they wanted to, but they haven’t. It is clearly not a technical issue — it’s policy.

Apple doesn’t want to promote a policy that leads to feature regression or data loss. Customers should buy new features, not pay a ransom for locked-down data, or blackmail money to get back features they already had. Practices like that leave a sour taste in a customer’s mouth, and I suspect Apple would rather avoid such tactics.

Paid Upgrades

Upgrades are already supported in many different forms. You can release a new version under a new bundle identifier, perhaps with introductory ‘upgrade’ pricing. You can have new features added as an In App Purchase (IAP), or have free and paid versions.

There is already enough flexibility in the system, and it is easy for customers to understand. Many developers have already issued successful upgrades in MAS, including skeptics like Delicious Monster and The Omni Group.

Introducing upgrade pricing to MAS seems like it should be a slam dunk, but dig a bit deeper, and you realize it introduces a variety of complications, potentially confusing for customers not used to the World of pre-App Store shareware.

To begin with, many customers would feel like they are paying for the same product twice — watch the one-star ratings flow in.

There are also design challenges: having to show a different price based on whether a user is signed in, and under what user name, could be confusing.

Issuing bug fixes to older, unsupported versions becomes challenging, and confusing to users. The developer would need to inject a release that precedes the most recent version. How such releases would be made visible to the user is not clear. There would probably need to be several buttons to distinguish between upgrading to the latest release, and just installing a free bug fix. In short, the MAS app could become quite a complex app, not the simple one-click to download affair that Apple favor.

Sandboxing

Most agree that Sandboxing is a good thing. I’m surprised that an app ever needs to be pulled from the store due to sandboxing restrictions, because Apple have shown a tendency to provide exceptions where warranted. If an app like Daisy Disk, which requires full access to your file system, can thrive in MAS, I don’t really see why nearly all apps can’t be made to work.

Reviews

As much as I dislike the review system, I’m not sure how much damage it really does. Most customers are probably astute enough to distinguish a considered review from a 1-star ranter. It would be nice if developers could contact reviewers offline, to help with support issues, but allowing developers to respond publicly would probably lead to ugly exchanges and wouldn’t do anyone any good.

Who’s at Fault?

If we accept that the Mac App Store is not living up to its full potential,1 who is to blame? For a while, I was inclined to blame Third Party Developers themselves. Until recently, I was of the view that developers were not seeing the potential in MAS, and were deliberately undermining it. Many developers do not even link to it from their web sites.

Actually, my view hasn’t really changed on that — I don’t believe that developers are embracing MAS, and that hurts the store. But my view on why that is the case has evolved. Now, I don’t blame Third Party Developers as much as I blame Apple. I think Apple made some fundamental errors with MAS, and that has hurt its chances of success from the beginning.

But the errors they made were not technical. I think MAS works reasonably well at a technical level, and the pricing models also work OK. So what did they do wrong?

Apple assumed the Mac software market was the same as the iOS market. The financial policies for developers in MAS are the same as for the iOS App Store. Developers give up 30% of every sale for the privilege of being in the store. On iOS, this is accepted, because there is only one channel to market. But on the Mac, there are other options.

When you pay less than 6% per sale to Fastspring, why would any developer actively promote MAS? Apple needed developers to push MAS, and most remained distant and unenthusiastic. When given the choice, developers pushed customers to secondary channels and pocketed 25% extra revenue for their trouble. Knowledgable customers would also choose off store options, to support their developers with a larger cut, and to avoid potential issues due to MAS restrictions.

With Developers actively undermining MAS, it is no surprise it has not lived up to its potential.

Where to from Here?

MAS does seem to be on a slow decline, and it certainly hasn’t become the one stop shop for Mac software that many hoped it would. What can Apple do to turn the tide now?

They can begin by acknowledging that they are dealing with a competitive market of resellers, and don’t have the monopoly they enjoy with the iOS App Store. They need to get developers enthusiastic about MAS, and actively promoting it. Dropping their cut to 15% would make it much more competitive and attractive to developers.

But I think they could go even further. We occasionally see apps pulled from MAS and become exclusively available via other channels. Apple should be trying to turn this trend on its head. It has used the stick of iCloud and other MAS-only technologies to try to ‘force’ developers into MAS exclusivity, but this has seldom borne fruit. I think a bit less stick, and a hint of carrot, could be the way to go: Apple could offer to drop its cut to 10% for any apps willing to be exclusive to MAS, and perhaps even introduce a ‘Mac App Store Exclusive’ badge.

Closing Shop

The combination of more developers pushing customers to use MAS, and more MAS exclusive apps, could be the mix that the store needs to reach its potential. Apple has levers it can pull, but must first acknowledge what the real problem is — apathetic developers who have little incentive to promote the technology.


  1. For every negative story about MAS, there are great success stories, so it is not clear if the store is actually in demise, or if only certain apps and categories are suffering. Apps like Sketch, Pixelmator and 1Password have boomed on MAS, and regularly compete head-to-head with Apple’s own apps for Top 10 Grossing list rankings.