How we made an App Store subscription success

Apple just announced they’re giving developers a 85/15 revenue split (instead of 70/30) for subscriptions lasting over a year — and Google will be offering even better terms.

As the co-creator of Zombies, Run!, a fitness app that transitioned to a subscription model just over one year ago, I couldn’t be more delighted. 🍾 + 🎉 all round, folks.

Before all that 🍾 + 🎉 though, I want to share the lessons we learned in the past year — a terrifying, exciting, and ultimately very successful year.

The Decline and Fall of Paid Apps

Once upon a time, my company, Six to Start, teamed up with lead writer Naomi Alderman to create Zombies, Run! It’s a running game and audio adventure where you actually run outside, in the real world.

We’re kind of a big deal, you know — we have hundreds of thousands of players and 3 million downloads. @wilw loves us!

We put Zombies, Run! on Kickstarter in September 2011, and we released it on time and on budget in February 2012. Would you believe that we charged $7.99 for the app? I know for a fact that other developers thought we were insane. But it worked out because we had a great app with zero competition and a fantastic concept.

One year later, we dropped the price to $3.99; the App Store was getting filled up with cheap or free apps, and users were increasingly reluctant to pay anything upfront. By 2015, things were looking ugly. We were still making a decent amount of money but we had fears for the long-term future of our paid app.

Something had to change.

Alone in the Dark

We didn’t know if switching from a $3.99 upfront payment to a free-to-play model would work. It’d eliminate one very high barrier to entry, meaning far more people would download and try out our exceedingly odd-sounding app, but it wasn’t clear how many of those people would convert. In fact, we had precisely zero experience at converting free users to subscribers.

Because Zombies, Run! is a fitness app, we couldn’t simply copy Candy Crush or Clash of Clans. Consumable microtransactions are perfect for addictive games that anyone can play any time, anywhere, but not for more specialised apps. And what kind of consumables would we sell? Power-ups that let you outrun zombies faster? That’s insane — we want our players to pay us for helping them exercise more, not less!

There were other subscription-based apps of course, like Adobe, Microsoft, Trello, or Spotify’s offerings, but they were huge and mostly business or content-led.

One of my Hampstead Heath runs, showing our web-based ZombieLink animated-paths feature

But we had some important advantages. Zombies, Run! had been maintained and improved for over three years. We‘d grown it from an MVP to iPhone and Android apps with enough story and gameplay for hundreds of runs, plus a web-based portal — ZombieLink, naturally. No-one could be in any doubt we were in this for the long haul, or that we were fully capable of making a high quality product.

And conceptually, subscriptions made a lot of sense. They aligned our players’ desires — exercising more over a long period of time — with our own. We would only get paid if they really enjoyed and valued our app.

The problem was, we’d have to increase the price.

Unleash the Shitstorm!

In May 2015, we made our iPhone and Android apps free and introduced a $3.99/month and $19.99/year subscription charge. Existing players received all content and features developed up to that point for free, forever, plus a discounted $7.99 ‘Legacy’ subscription.

New players could access the first four missions of the game with no limits, and then would be able to unlock one mission per week for free. They’d hear audio ads before missions, and they wouldn’t have access to neat features like Airdrop mode, Interval Training, and zombie chase controls, but otherwise they could play through everything we’d made for free. Here’s lots more details.

Three things swiftly occurred:

Combined iOS and Android downloads in May/June 2015, following our free-to-play transition.
  1. Downloads soared. We had hundreds of thousands of new users in the space of a month.
  2. After a burst of new Legacy subscriptions from our loyal fans, we stopped making money. 💩.
  3. A shitstorm erupted, mostly on our subreddit, in which we were accused of being ‘greedy devs’. Attempts to politely engage were consistently rebuffed. It was incredibly stressful and disheartening.

And then one thing happened slowly but surely: new players began to subscribe, mostly at the pricier but more cost-effective $19.99/year. That shouldn’t be surprising since most people aren’t dumb — and anyway, if you’ve decided you like an app enough to pay for it, you might as well commit to a longer period (especially if it’s a fitness app).

But it was lovely to see, and they kept coming.

Obligatory ‘Five Steps to Success’

This is Medium, right? Thought-leaders gotta lead thoughts.

1. Be Reliable

Everyone hates committing to new subscriptions. You live in fear of signing up and then the service instantly becoming unusable. Not only are you disappointed, but even worse, you’re a rube.

So you need to do everything you can to reassure your users that you’re in for the long haul. That means regular, consistent updates and bug fixes. You don’t need to release a new build every two weeks like Facebook, but you need to demonstrate commitment to maintaining a stable and reliable app — one that adopts useful new features (e.g. Healthkit, Apple Watch) in a reasonably timely manner.

This is the opposite of a big bang release once a year, laden with new features and new bugs. Frankly, it’s a much more sustainable, relaxed, and considered mode of development. It means you can justify the time to achieve 99.9% crash-free sessions, as we’ve done.

2. Solid Customer Support

Everyone hates doing customer support. Even if you receive 99 positive, easily-solved tickets in a row, that one asshole ticket can ruin your day. Believe me, I’ve answered thousands of tickets myself.

But customer support is crucial because it helps you identify problems, and it really improves word-of-mouth. And it doesn’t need to take up all of your time. After we made Zombies, Run! free-to-play and our active users rocketed, in the long term our support load actually reduced.

Average support ticket first reply time in hours

Why? Because we spent months improving UX so people didn’t have to ask us how to do things, and we created an extensive in-app and online knowledgebase. The result is that we answer people’s tickets within 24 hours.

3. Steady Improvements and Content

I’ll grant that Zombies, Run! is hardly the easiest app to learn from. One area in which we’re different is our incredible, dramatic, thrilling audio story written by Naomi Alderman and directed by Matt Wieteska.

We release new missions every week, and we have over 200 in total. That’s a big reason why people start subscribing and keep subscribing — they want to hear what happens next, and they aren’t willing to wait a whole week to unlock for free.

And yes, we add new features — but not as many as in the past, because Zombies, Run! is a very mature app at this point. We spend a lot of time adopting Material Design on Android, supporting new OS-level features like HealthKit, the Apple Watch, and in the future, Google Fit and Android Wear. These make our app more valuable, and they demonstrate our commitment.

4. Extensive Trial Period

No-one can tell you how much you should offer for free or for how long, and even A/B testing has its limits. Maybe we should offer more than four missions for free; maybe we should let people fewer missions per month.

The important thing is that new users have to understand the value your app holds, and they can only do that by trying it out, sometimes for a long time, if they use it infrequently.

5. Explain the Benefits

Let’s face it — most users don’t read the emails or blog posts you write very closely. Even so, you need to be crystal clear about why you are inevitably charging more for what appears to be exactly the same product.

We explained that subscriptions would allow us to maintain and improve Zombies, Run! for many years to come.

We explained that subscriptions gave us independence from VCs, investors, and acquirers looking for a quick buck. Six to Start was bootstrapped; no-one is looking for a 10x return. I personally find it very rewarding to make a game that improves people’s lives every single day — a game that has helped players recover from cancer, stop self-harming, and brought together families. But we can’t do it if we’re starving.

We explained that $19.99 is barely the price of a two weeks’ gym membership. It’s just 5 cents per day! (hint: no-one cares, but you’ve gotta say it anyway).

And we explained that we’d only gotten here because of the support and loyalty of our paid users, who we were in turn rewarding with a massive and permanent amount of free content, features, plus a discounted legacy subscription.

And our results show that I think the message was received and understood.

The 85%

As our first annual subscribers on iPhone begin rolling around, I’m obviously very grateful that Apple has effectively given us a 20% boost on renewals, which we can plough back into development (we also love the 48 hour review times!). I see it as a recognition that we’ve built an incredibly valuable app that benefits Apple as much as anyone else.

Let’s be clear though — subscriptions are not easy to implement, especially if you’re making a new app. You need a real stack of cash in the bank to help you survive through the lean early days. But what Apple’s done is make it easier for successful subscription-based apps to survive and thrive.

It’s a big step in the right direction.

Post-race selfies from the 2015 Zombies, Run! Virtual Race

If you found this article useful, please check out Zombies, Run! on iPhone and Android, or our super-cool Zombies, Run! Virtual Race, coming this October.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s