Since publishing my piece on revenue distribution in the App Store yesterday, I’ve gotten some feedback from readers asking how valid my extrapolated data is. Some have pointed out that I’m working with a limited data set from only one app that might not hold for all apps. Others have pointed out that the daily revenue data that Marco Arment provided did not include any data from the period when Overcast was at the top of the U.S. Top Grossing list, so it’s possible that the curve isn’t really a strict power curve all the way up the charts.

These are valid criticisms. So let me respond by releasing more data; the only other useful data that I have available. In response to Marco’s post, William Wilkinson published financial data from his own app, Manual. Manual is a photography app that did well on the U.S. Top Grossing list (reaching as high as number 104) in September 2014, so the Manual revenue data is both contemporaneous with Overcasts’s and useful for examining the upper reaches of the App Store. The problem with the data Wilkinson released is that the daily revenue graph has a very low resolution. In reading the graph, I could estimate daily revenue to only ±$133. That’s not such a big deal at the high end, where daily revenue is measured in the tens of thousands of dollars, but in the long tail where revenue is measured in the hundreds of dollars, a $133 error is quite significant. With this limitation in mind, below is the graph of Manual’s data overlaid onto Overcast’s.

The Shape of the App Store, with Manual Revenue

As you can see from the graph, the revenue curve for Manual breaks upwards a little before Overcast’s does, but overall they are very similar curves. In particular, notice that the “head” end of these graphs (U.S. Top Grossing #1 to about #1000) follow similar paths. It seems likely to me that the differences between the graphs can be accounted for by the daily fluctuations of the App Store, differing ratios of U.S. to global sales, and the inherent inaccuracy that comes with estimation.

So what does this mean? Well, it makes me think that I was on the right track with my original post. While the the exact revenue figures that I cited in The Shape of the App Store may be off little, it seems likely that they are at least in the right ballpark. They may even be somewhat conservative estimates if Manual’s revenue figures are to be believed. Whatever the exact figures, one thing is clear: Revenue is heavily skewed towards the top of the charts, to a much higher degree than I, at least, expected.

Posted on January 20, 2015

Read More

Every developer knows how tough it can be to make a living on the App Store. There’s a lot of money being made there, but it’s not spread very evenly. Those at the top of the charts make the lion’s share of revenue, while the vast majority are left to fight over the scraps. But exactly how lopsided is it? And how does that affect an indie developer’s chances of finding success? For a long time, I was resigned to never really knowing the answers to these questions, because although there’s wide consensus that the distribution of revenue in the App Store is best represented by a power law, I had no way of knowing how sharply or gently the graph of that power law curved. My own apps haven’t spent enough time on the Top Grossing list to collect the data I would need to make a prediction, and Apple sure isn’t sharing it.

And then on January 15, Marco Arment gave me (well, all of us) a belated Christmas present. In his piece titled Overcast’s 2014 Sales Numbers, Marco shared the financial performance of his podcast player, Overcast, through a series of statistics and graphs. Because the information he shared was so detailed, and because Overcast has been so successful, I realized that Marco’s post could provide some insight into how an app’s rank on the U.S. Top Grossing list correlates with its daily revenue.

Before we get to the results, it’s important to first discuss their limitations. The figures that Marco provided for Overcast were very detailed, including daily revenue information, but they weren’t complete. Of particular importance, the revenue figures that Marco published were for global revenue. There was no breakdown of the revenue provided by nation. This means that it’s possible only to compare rank in the U.S. App Store to global revenue. It’s not a perfect apples to apples comparison, but since the U.S. App Store is the largest national App Store both by downloads and by revenue, I think it’s a reasonable comparison to make.

To provide some context to the results, you may be familiar with the Pareto distribution. It’s the origin of the classic “80-20 rule” that’s used to explain so many phenomena that obey a power law. “Twenty percent of the people in an organization do eighty percent of the work.” “Twenty percent of the population control eighty percent of the wealth.” You hear these types of statistics a lot, but they’re usually not very accurate. Often, they are useful as a first estimate at best. So I didn’t actually expect App Store revenue to obey the 80-20 rule. In fact, I expected it to be a much sharper curve, representing even greater disparity in the distribution of revenue than the 80-20 rule would suggest – maybe a 90-10 split, or even a 95-5 split. As it turns out, the revenue distribution curve of the App Store is even sharper than I imagined.

The Shape of the App Store

I expected a “hockey stick” curve that’s characteristic of power law models, but I didn’t expect one like this. The hockey stick breaks upwards at around position 870 on the U.S. Top Grossing list. With about 1.2 million apps in the App Store at the time the data was collected, that arguably puts 99.93% of apps in the “long tail” of the App Store. The “head” of the App Store, those 870 top grossing apps that make up 0.07% of the App Store population, collect over 40% of the App Store revenue that’s paid out.

Luckily, there’s a lot of money to be made in that long tail. At the top of the long tail, in position 871 on the U.S. Top Grossing list, an app still makes over $700 in revenue per day. That’s almost $260,000 per year. Even number 1,908 on the U.S. Top Grossing list makes over $100,000 per year. In fact all apps above number 3,175 on the U.S. Top Grossing list produce enough revenue to at least make its developer the United States household median income for 2014 ($53,891). And this is just for a single app. Most indies I know develop more than one app simultaneously. Developers who can put together a collection of apps that rank at about 6000 on the U.S. Top Grossing list (about $25,000 in revenue per year) stand a good chance of building an app business that can sustain them and their families.

Now, all of that is not to suggest that it’s easy to build a sustainable app business. It’s not. But, by doing some basic business things right – like finding a profitable niche, solving a painful problem, effectively marketing apps, and generally doing the things that have been suggested time and again at conferences and on podcasts – a developer can dramatically increase his odds of out performing the multitudes that just slap any old thing up on the App Store and hope for the best.

So, with even fewer people than I expected making “yacht and helicopter money” in the App Store, I remain hopeful for my fellow developers. There’s a lot money circulating in the ecosystem, and a developer operating at indie scale only needs a little bit of it. It seems that even with the revenue curve tilted so heavily towards the big hits, the shape of the App Store still allows room for sustainable businesses to develop in the long tail. It seems that developers who work hard, mind the details, and treat their business like a business have a real chance of making it.

Note: I’ve posted a follow-up to this article titled, The Shape of the App Store, Redux that responds to some valid criticism that’s been received regarding the data set used here.

Update Jan (20° 20°)15: This post was updated to clarify that an app’s rank in the U.S. Top Grossing list is being compared to its global revenue.

Posted on January 19, 2015

Read More

In episode 206 of his podcast Developing Perspective, David Smith talks about the idea of the App Store being “full.” He makes the argument that with almost 1.5 million apps in the App Store, there are now enough apps to choose from and that all the good ideas have been taken; that for any basic type of app that you’re likely to think of, an app already exists to fill that need. I agree in large part with David’s observation that the App Store is now full, but the App Store being full is, I think, merely a side effect of something more important: The App Store has matured.

Yes, the shelves of the App Store are now fully stocked, but you’d expect to find that in Target or Tesco, or any mature market. What’s more significant is that the shelves of the App Store, more than ever before, are stocked with products from bigger players that compete at a higher level. These big players have more time, more money, better connections, and more collective talent than any single independent developer. They have the resources it takes to compete in the mass market that the App Store has become. These big players have money. They have reach. They can create mindshare and build valuable brands in a way that most indies just cannot afford to. The sad reality is that we indies generally can’t stand toe to toe with large corporate developers and hope to win. And so we shouldn’t try. Let the big guys duke it out over most of the market. We don’t need the whole pie. We just need a sliver. We need just enough to live our life, pay our bills, and maybe save a little bit too.

To get our sliver of the pie, though, we need to adjust our strategy. As David correctly points out, novelty is no longer a viable business plan. In an App Store that’s full, indies can’t depend on new ideas to attract the attention and the audience we need. Instead, we need to go where the big players aren’t. We need to compete in niches, where there isn’t enough opportunity to justify the attention of large corporate developers. Don’t try to create a new bookkeeping app – Intuit will eat you alive. Instead build a bookkeeping app that’s tailored specifically for veterinarians or, even more narrowly, for large animal veterinarians. Don’t build a general purpose word processor – Microsoft has that space all locked up. Instead, build a word processor that’s specialized for a particular field like academics or screenwriting. Each of these niches offer plenty of revenue opportunities for a single developer. The big players won’t be interested, though. After all, a niche with potential annual revenue of $250,000 might be an amazing opportunity for an indie, but for the big players, $250,000 won’t even cover their engineering costs.

By competing in niches where the big players don’t want to be, indies level the playing field. They avoid going head to head against large companies with massive resources. They avoid struggling against slick advertising campaigns for the competiton. They avoid having to compete with venture capital backed companies that give their products away for free. By competing in niches and avoiding match-ups that they can’t hope to win, indies reduce both the quantity and the quality of the competition that they face in the App Store. And that gives them a sort of freedom. The freedom to seek out only the most valuable customers. The freedom to charge sustainable prices for their work. The freedom to stay profitably small.

The real beauty of this strategy, though, is that it means, for indies, it doesn’t matter if the App Store is full. There’s always room for another niche product. Just as there’s always room for sand to squeeze into the gaps between large rocks, there will always be room for indies to enter markets that aren’t big enough – aren’t mass market enough – to support large companies with teams of developers. It’s in these gaps between mass markets where indies can make their living. It’s in these niches, within an already full App Store, where indies can find success.

Note: I expanded on this topic in a talk I gave at NSNorth this year. NSNorth has posted the video of that presentation online.

Posted on December 16, 2014

Read More

As you’ll hear on this Monday’s episode of Release Notes, some of the new features of iOS 8 have me wondering about the future. Based on what we saw at WWDC, it seems pretty clear that Apple intends to introduce devices this fall with screens of different dimensions. The introduction of size classes and other features all point in this direction. The rumor mill has also been hinting at larger iPhones for some time. In particular, rumors have been flying about a new iPhone that will come in both 4.7-inch and 5.5-inch sizes. No one knows for sure what Apple has planned, but at this point I think the rumors of a larger phone are likely true. Assuming that’s the case, what might that mean for the iOS ecosystem?

iPhone and iPad size comparisonThe most obvious change, as this image from MacRumors makes clear, is that we would no longer be dealing with two distinct device sizes as we have to this point. Instead, the addition of a 4.7-inch and 5.5-inch iPhone would blur the lines between “iPhone-size” and “iPad-size” devices. We’d have a spectrum of sizes, with the 5.5-inch iPhone almost the size of the iPad mini. In fact, a 5.5-inch iPhone in landscape mode might actually be wider than an iPad mini in portrait.

[pullquote align=”right”]”I find it entirely plausible that a 5.5-inch iPhone in landscape would have a horizontal size class of ‘Regular,’ just as the iPad presently has.”[/pullquote]This brings us back to the new iOS 8 size classes. (If you’re not familiar with iOS 8 size classes, you might check out this post by Justin Williams for a primer.) I find it entirely plausible that a 5.5-inch iPhone in landscape would have a horizontal size class of “Regular,” just as the iPad presently has in both orientations. We’ve already seen that a 5.5-inch iPhone would be almost the same physical width as a portrait iPad mini, and notably, both rumors I’ve seen for a 5.5-inch iPhone would provide more horizontal pixels for a landscape 5.5-inch iPhone than exist in today’s portrait iPad. (The screen width of an iPad in portrait is 1536 pixels. The width of a 5.5-inch iPhone in landscape mode is rumored to be either 1704 pixels or 1920 pixels, depending on who you believe.)

As the lines between “iPhone-size devices” and “iPad-size devices” blur, interface conventions might blur as well. Taking the assumed 5.5-inch iPhone as an example, it might have a “compact” horizontal size class in portrait mode, and therefore conform to the interface conventions that we today associate with the iPhone (e.g. single table view presentations, no popovers). In landscape, it might have a “regular” horizontal size class and take on some interface conventions that we today associate with the iPad (e.g. split view controllers with both a master and detail view). In short, the bright line that has always existed in the minds of developers between an iPhone interface and an iPad interface may start to become not so distinct. And this is where things start to get really interesting from a business perspective.

Since the introduction of the iPad, Apple has consistently pushed developers to create “Universal apps,” or apps that run on both the iPhone and the iPad with a single purchase. A lot of developers view that as leaving money on the table, though. Today in the App Store, it’s not uncommon to find the same app present in both iPhone and iPad flavors that are available for separate purchase. This is especially true in the Productivity category, where top apps such as OmniFocus, Things, Launch Center Pro, and even my own apps are available with separate versions intended for the iPhone and iPad.

Having separate iPhone and iPad versions makes perfect sense from the perspective of a developer. It takes effort to provide a great experience on both the iPhone and iPad. Why shouldn’t we be compensated for the extra value that we provide? Contributing to that line of thinking has been the clear dichotomy between the iPhone and the iPad. Up until now, the iPhone interface was seen only on the iPhone, and the iPad interface was seen only on the iPad. As a result, it was a fairly straightforward to explain to customers why they had to purchase an app twice: “They’re two different devices.”

[pullquote align=”right”]”As the lines between the iPhone and iPad start to blur, the argument becomes harder to make that an app should be a separate purchase on the iPhone and iPad.”[/pullquote]But as the lines between the iPhone and iPad start to blur – as we start to see interface elements appear on the iPhone that heretofore were considered iPad interface conventions (e.g. a split view controller in a landscape 5.5-inch iPhone) – the argument becomes harder to make that an app should be a separate purchase on the iPhone and iPad. In fact, I think that it makes it harder to argue that there is such a thing as an “iPhone app” or an “iPad app.” If an app on the iPhone has to be ready to present a split view controller, is it really “an iPhone app?” If an app on the iPad has to be ready to collapse itself down to a single table view when displayed in the rumored split screen mode, is it really “an iPad app?” Or are they instead, just iOS apps?

Developers who currently sell separate iPhone and iPad versions of their apps would be well advised to start thinking about how they plan to handle the blurring of lines between these devices. If your customers no longer see a difference between the iPhone and iPad, how will you justify to them the need for separate purchases? If Apple no longer sees a clear dichotomy between the iPhone and iPad, might they no longer accept app updates that are not Universal? If you find yourself in the position of having to give away what you once were paid for, how will you make up the lost revenue? I don’t have answers to these questions, but I’m convinced that I need to find some. Maybe sooner than I’d like.

Posted on June 25, 2014

Read More

At last week’s Worldwide Developer Conference (universally referred to as WWDC), Apple gave independent developers a lot to be thankful for. In fact, Apple gave us many of the things we’ve been requesting for a long time – things like app bundles, preview videos in the App Store, better options for beta testing, and App Store analytics. One thing that Apple didn’t give us, though, is upgrade pricing, or the ability to charge owners of our current software a reduced price for the next major version of that software.

Upgrade pricing is a staple of the non-App Store software market, and its absence from the App Store has long been considered one of the biggest obstacles to creating a sustainable software business within Apple’s ecosystem. Wil Shipley wrote the seminal piece on the subject back in 2012, but complaints about the omission of upgrade pricing in the App Store go back much further. It seems clear at this point that Apple has no desire to add upgrade pricing to the App Store, so conventional wisdom has been that developers had better adjust to the new reality and get on with the business of building software.

But Apple may have unintentionally cracked the door to upgrade pricing with one of its announcements last week. In the latest episode of Release Notes, you can hear my podcast co-host Joe Cieplinski wonder aloud at the possibility of using app bundles as a form of upgrade pricing. His comment was off-the-cuff, so the idea wasn’t fully developed, and we both quickly dismissed it, but upon further consideration it appears that it might be possible to simulate upgrade pricing through the use of app bundles.

First, a little background. In the WWDC Keynote, and later in Session 302: The New iTunes Connect, Apple announced that they will soon allow developers to create bundles of their apps. The gist of the information provided is that developers will be permitted to bundle up to 10 of their apps and make them available to be purchased as a group for a reduced price. One of the key features announced about app bundles is that consumers will be able to complete a bundle at a “discounted price” if they already own some of the apps in the bundle. It was never defined what exactly was meant by a “discounted price,” but it seems reasonable to assume the simplest case – that customers will just have to make up the difference in price between the cost of the bundle and the cost of the sum of the individual apps that have already been purchased. For example, if an app bundle costing $14.99 contains two apps, App A (priced individually at $9.99) and App B (priced individually at $9.99), then a person who already owns App A could “complete the bundle” and receive App B at a cost of $4.99.

But what if instead of bundling two different apps, we instead bundled two different versions of the same app, released as two different SKUs? Modifying the example above, what if our $14.99 app bundle contained My App (priced individually at $9.99) and My App version 2.0 (priced individually at $9.99)? That would allow people who owned My App to purchase the new version 2.0 at the reduced price of $4.99, while still requiring new customers to purchase version 2.0 for full price. This is exactly the effect that we have been seeking with upgrade pricing all along!

Using app bundles to simulate upgrade pricing would not be without its pitfalls, though. Bundling two versions of the same app would probably require that the old version remain available for sale individually. This would inevitably lead to customer confusion and probably cause some customers to purchase the wrong item. It seems like this problem could be alleviated, though, by being generous with promo codes (or app gifting) with customers who contact you with problems. Nonetheless, it’s probably desirable to limit the time that the app bundle is available so that the original version can in a timely manner be removed from the App Store and officially discontinued.

While I still hold out hope for genuine upgrade pricing in the App Store, I realize that it’s unlikely to be introduced any time soon. Until that time, app bundles appear to offer many of the advantages of upgrade pricing with manageable disadvantages. I’m hopeful that Apple will not take action to prohibit this type of bundling, and that this type of simulated upgrade pricing will allow more complicated, higher priced apps to find a sustainable market in the App Store. So thank you Apple!

Posted on June 9, 2014

Read More

The Alt Alt Dub Dub schedule for Friday, June 6 is below. If you’d like to join us in the Parc 55, please contact Charles Perry by email at or on Twitter at @DazeEnd. In order to offset the cost of renting the room and equipment, there will be a $25 per day charge for admission. Sessions are already showing, so come on over.

9 AM 409: Introduction to LLDB and the Swift REPL
10 AM 406: Integrating Swift with Objective-C
11 AM 404: Advanced Swift
Noon Lunch Break
1 PM 226: What’s New in Table and Collection Views
2 PM 223: Prototyping: Fake It Till You Make It
3 PM 304: Creating Great App Previews
4 PM 411: What’s New in Interface Builder

Posted on June 6, 2014

Read More

The Alt Alt Dub Dub schedule for Thursday, June 5 is below. If you’d like to join us in the Parc 55, please contact Charles Perry by email at or on Twitter at @DazeEnd. In order to offset the cost of renting the room and equipment, there will be a $25 per day charge for admission. Sessions are already showing, so come on over.

9 AM 214: View Controller Advancements in iOS 8
10 AM 403: Intermediate Swift
11 AM 302: The New iTunes Connect
Noon Lunch Break
1 PM 221: Creating Custom iOS User Interfaces
2 PM 216: Building Adaptive Apps with UIKit
3 PM 212: Storyboards and Controllers in iOS 8
4 PM 406: Integrating Swift with Objective C

Posted on June 5, 2014

Read More

tl;dr: Come join us in the Parc 55 to watch WWDC sessions as they they are released. Space is limited and there is a small daily fee to offset the cost of renting a room.

I came to WWDC 2014 without a ticket, as did a lot of other developers this year. Even if you don’t have a ticket, just being in San Francisco for WWDC is a great time, but not having a ticket makes it tougher to watch all the sessions you want to watch. Apple addressed this issue last year by releasing the videos of many WWDC sessions the same day that they were presented. But it’s no fun to hole up in your hotel room and watch technical videos. Alone, you miss out on the discussions and the shared insights of your fellow developers. Because I and a group of friends wanted a more social experience while watching videos, I rented a meeting room in the Parc 55, arranged for a projector, screen, and audio, and dubbed the result Alt Alt Dub Dub.

But yesterday, Apple made a big improvement to our WWDC experience. They dropped the non-disclosure agreement on WWDC sessions. That means that we can discuss what we learned in WWDC without legal worries hanging over our head. What that means for Alt Alt Dub Dub is that I don’t have to be as careful about who joins us for WWDC sessions. And so we are opening our doors to other ticketless developers who happen to be in San Francisco this week.

Space in our meeting room is very limited, but if you would like to join us to watch WWDC sessions as they are released, we would love for your to join us. In order to offset the cost of renting the room and equipment, there will be a $25 per day charge for admission.

We expect seats at Alt Alt Dub Dub to be in great demand. If you want to reserve a seat, please contact me by email at or on Twitter at @DazeEnd. We’re showing sessions starting right now, so the sooner you sign up, the sooner you can join us for sessions.

Our schedule for Wednesday, June 4 is:

9 AM 202: What’s New In Cocoa Touch
10 AM 402: Introduction to Swift
11 AM 401: What’s New in Xcode 6
Noon Lunch Break
1 PM 211: Designing Intuitive User Experiences
2 PM 205: Creating Extensions for iOS and OS X, Part 1
3 PM 209: Adapting Your App to the New UI of OS X Yosemite
4 PM 208: Introducing CloudKit

Posted on June 4, 2014

Read More