Full Analysis of iPhone Economics: Bad News?
communities-dominate.blogs.com
communities-dominate.blogs.com
I'm one of those developers. And do you know why? Because it's not my business. I wanted to make some apps. I didn't put in a huge effort.
For the numbers to mean anything, you should filter out the 90% (or whatever) of apps that aren't intended to be part of a serious, profitable business.
He doesn't address something that maybe you're touching on, indirectly - unlike the markets where someone makes zillions on SMS voting for a popular TV show or selling Tetris to feature phone users, the iPhone app market is very accessible to J. Random Hacker. Compared to the previous age of mobile development, it's a major improvement - same device (more or less), same screen size (more or less), same framework, no complicated permission needed from the carrier, payment taken care of, etc, etc. Plus, you never hear of anyone saying they program J2ME or BREW or Symbian apps for fun.
No, 3 of my 4 apps are paid, even though I'm not making a serious effort to make money. Why shouldn't they be? I don't mind making a little money.
Just because filtering the numbers to get useful results is infeasible doesn't mean we should pretend averages across the entire store mean anything for a person who sets out to build an app store oriented business.
It reminds me of all the psychological studies that use inadequate controls just because adequate ones for what they want to do would cost trillions of dollars. Then they claim to have done legitimate science. But the standards of science don't lower themselves to what data we can easily get.
Or at least one should carefully document every single wild guess being made when coming up with some numbers, and carefully identify them as very plausibly being very large sources of error (so that no readers mistakenly doubt that the numbers might be off by a factor of 10 and think it's actually based on hard data).
Nobody said they shouldn't be. The point is 'serious effort' vs anything else is not readily measurable. This is something you say, rather than something you do. 'Trying to make money' vs 'Not trying to make money' is measurable - if you're not trying to make money, you don't charge. If you charge, you are obviously trying to make money, even if it's very little money.
Just because filtering the numbers to get useful results is infeasible doesn't mean we should pretend averages across the entire store mean anything for a person who sets out to build an app store oriented business.
The numbers are still useful, whatever their caveats and limitations. At the very least, they are more useful than non-numbers or numbers-that-ought-to-be-obtained-by-some-method-that-doesn't-actually-produce-measurable-numbers.
But the standards of science
I don't think there was any claim to reaching the 'standards of science'. It seems a little silly to bring them up, though - the author's sources and methods are well described. There's probably plenty of room to take issue there. But you can't trot out 'the standards of science' if your standard is some underivable set of data that just happens to support your viewpoint.
Today if you decide to develop a free app, you pay your $35,000 development cost, then you find your free app competing against just a modest set of 60,750 rival free apps on the iPhone. Imagine if you are your consumer, scrolling past that set, to find you.
Firstly 35k assumes a decent sized app, not a few day job like a lot of these apps are. Secondly a lot of these apps have NO rivals. Search 'Nottingham Cinema times'. 2 apps. But guess what. Those guys also make 'Leicester Cinema Times', 'Birmingham Cinema Times', 'London Cinema Times', etc., etc.
All the same tiny piece of code.
And what about all the hoo-hah about those guys who had 20,000 apps like 'Pamela Anderson Pictures' that were eventually pulled? Guess what. Search 'Claire Danes', 'Kiera Knightly', any female actress. 59 p app for each one of them. Probably thousands of them. Same piece of code.
And look at some of the numbers he throws about, 2.1 billion app capable phones vs. 80 million iPhones. I'm sorry, but add up ALL of Europe and ALL of North America and you've not got 2.1 billion people. But I bet a large proportion of those iPhones are in that market. And I bet the vast majority of app purchases are in that market.
And worst of all for those 2.1 billion phones. THERE'S NO MARKET.
Roughly 50% of the people I know have an iPhone. And I'm probably downplaying that number. Young professionals with large disposable incomes. Just the sort of people you can make a lot of money off.
He's not comparing like for like, he's not realising that the median earnings per serious software shop is actually higher much higher than the average per app because of so much dross and he claims a 'typical' app costs 35k without considering that they probably only asked the serious software shops and not people like the poster above you or those shops which are pumping out 100s of apps per week.
Well, you can keep saying that but you have no numbers and no particularly concrete criticisms of his argument. Maybe he's not comparing like for like. Maybe all your friends have iPhones. Maybe you've forgotten many people outside of North America and Europe own mobile phones. The $35k, as mentioned in other comments in the thread, has very little bearing on his central argument that iPhone apps don't make much money.
And maybe he's made a complete hash of it. But at least we know how he got there.
he's not realising that the median earnings per serious software shop is actually higher much higher than the average per app because
Because what? What's a 'serious software shop'? How do you come to this conclusion? Right or wrong, I have a reasonably good idea how the author of the article came to his. Can't say the same for yours.
I wonder how many of those folks have phones that allow them to participate meaningfully in an application marketplace.
Science is how we know things. This article doesn't help us know very much, because it's so sloppy with its numbers.
I don't know how you would split out Brand apps vs hobbyist vs junk apps, but he should have mentioned that. I'd be surprised if there are more than 10K Brand apps like the ones he appears to be most familiar with.
If someone is churning 20 in a month and then getting 1364$ * 20 = 27,280$ over two years for one months work that's not bad.
PS: The average app has probably only been out for 1 year so it may be reasonable to double these numbers. Basically, anything released yesterday is unlikely to have generated much revenue, averaging ALL apps over a 2 year earning period is silly.
This guy makes a strong case for skating to where the puck used to be.
All iphones are SMS-capable. Not all SMS-capable phones are iphones. Just like Twitter never killed blogging, smart phones still haven't killed plain old dumb phones, either.
It is pretty funny that the author includes WAP as one.
I agree, the chance of one app being the runaway success that affords one an early retirement is pretty remote, but the tech press loves to promote the huge success stories, and many others love to follow them.
You can treat apps the way the music industry treats songs with the ambition that 1 in 100 is a "hit", and cover your losses with the success of the few...
...or you can provide a unique app that serves an existing need and charge a reasonable price (far above the typical app store prices).
There are countless examples of the latter, and not only are they successful financially but they serve to raise the standards of quality and compensation for the other type of participants.
One of my friends met the developer of Air Horn. The #1 downloaded free app for endless weeks. He said the guy had over three hundred apps on the App Store until he hit it big. Really big.
http://itunes.apple.com/us/app/air-horn-free/id348184873?mt=...
That is hardly a blanco card for writing apps in Lua. Anyway - not my problem.
First off, in app purchases are not included. Secondly, his argument against free apps is simply that the reach is smaller than other platforms. Finally, he acknowledges that the median paid app get fewer than 1000 downloads.
The 25-50K number for development costs is way off for the median app. Something like $5K-$10K is more in line for most of the apps I'm seeing (but I work cheap), and I've done a couple small ones (probably more representative of the bottom half of the distribution) for clients in 3 days' time.
Simply saying that targeting an iDevice leaves out 97% of the total population might be meaningful to an advertiser trying to gain a wide reach, but it means less to a developer, who is mostly interested in 1) getting an app installed on the device and 2) getting impressions. The fact that the average iPhone user has downloaded 80 free apps is a good thing for a developer.
Let me know how much money you make doing iPhone apps with AdMob. I'll be here.
So maybe developers are advertisers, but they are not buying ads, they are selling ad space in their app. The numbers for iPhone and iPod Touch devices from last September should be interesting to developers (http://techcrunch.com/2009/11/09/admob-is-approaching-100-mi...):
In September, AdMob served up 10.2 billion ad requests across 15,000 mobile Web sites and apps. About 28 percent of those were from iPhones and iPod Touches, with Android growing fast, making up another 7 percent. And if you look at only smartphones, Apple devices accounted for 48 percent of ad requests, with Android coming in second at 17 percent.
That's nearly 3 billion iPhone/iPod impressions in a month, just from AdMob. There are some key pieces of missing information - like the breakdown of the 15,000 request sources, especially web app versus native app - that would enable some more interesting analysis. But just saying that the iPhone reach is small doesn't really address the usage patterns of iPhone users versus other smartphone users.
Maybe you can't make decent money on ad-driven apps. I don't know. I haven't tried. The closest I can come to any information are a few anecdotes from usenet posts that claim anywhere from 10's of dollars to 100's of dollars a day in ad revenue. Not useful at all. But that doesn't make saying that iPhones are only 3% of the market a "full analysis."
As a sole developer, you require not only the technical knowledge to complete the initial job of writing the software that attracts the traffic to generate revenue, but you also must rely on yourself to place and promote it adequately to have significant returns enough to cover both investments of time and energy.
"80 apps" is great news for a developer if a good percentage of them are yours, serving your ads. Otherwise, its at best 1/80th of a chance they are running your ads, but probably worse as each app is not created equal. If yours is a "killer app" that is use most often, its more than likely not ad-driven unless a guy likes playing iChess on the shitter.
All of the data you provided is great - but none of it ties to money paid. 10.2 billion ad requests means dick if there wasn't any money exchanging hands. You said yourself there's no definitive evidence of people making a lot from mobile app advertising, so I'm not sure what point you were defending in the first place.
Anyone who has used iPad or iPhone or an Android device can tell that devices like these, not Symbian or WAP or Brew or whatever, are the future. Those other platforms might catch up, but for now they lag abysmally behind in terms of user experience. And so while they might outnumber iOS and Android for now, that numerical advantage can not last.
Walk around a Y-Combinator event (or CES, or E3, or anywhere else where leading edge alpha geeks congregate) and count phones. I'd be willing to wager a lot of money you'd see a bunch of iOS and Android devices, a few blackberries, and a smattering of ancient nokia and samsung free phones held by users who "just want to make calls" (and thus aren't going to be reachable by any apps at all, WAP or Java or otherwise).
So, yes, a company developing an app today for iOS or Android isn't going to make their money back on the platform as it stands in any reasonable amount of time. But a company or developer who positions themselves to be on top of a platform that is rising meteorically (and will displace literally billions of feature phones over the next decade and a half as moore's law drops silicon costs by a factor of 1000) is going to do well.
Hype is about future expectations. When expectations rise above exponential growth in our industry, those expectations usually turn out wrong. But the mobile market has a current penetration of over 3 billion people and I'll wager practically every phone on the planet will be a "smartphone" in under 15 years. That's a lot of headroom for growth.
But, don't we say that only one in ten start-ups makes it? So, as a baseline, why expect more than one in ten apps on the App Store to be profitable?
Therefore all of his numbers and his final conclusion are wrong. People do not write iPhone apps in order to achieve some sort of 'average' income. They hope to strike it lucky and get a disproportionately large payoff.
"This is the average, remember, it is not the median. The average skews too high because of the long tail. There are a few who make several millions, who distort the average number, so it is not true that half of iPhone App Developers earn more, and half less, than $3,050 per year."
He references the long tail you mention, and describe exactly what you mean about some people making a lot, but because of distribution and how users find and download apps - its more akin to say - the lottery - than actual market strategy.
Some will make money. Most of them will be development/contractor shops that charge out-the-ass for mobile apps to companies who want a "mobile presence." These medium to large companies won't realize that the money they are losing isn't worth it until its happened a couple of times. Then they'll stop coming back.
That said, independent developers making money on that app store directly is a complete crap-shoot. Very few get rich, lucky ones make a very modest living, more get a hobbyist return, most get dick - and some don't even get the money they did earn from Apple.
Therefore all of your rebuttal and final conclusion are wrong. How many books have you written on the subject?
Note: The reason he uses a normal distribution is because it would actually help DISPROVE his thesis, a normal distribution would make it easier for an app developer to make money, not harder. When the normal distribution showed a shitty return - he reminded us it was actually worse.
The reality is that taking an average over an exponential distribution is largely meaningless. There are an infinite number of distributions that would create that exact mean. Some of them not at all profitable for certain apps, others very profitable.
The problem with using an average over this distribution is that it doesn't give you anything to work from. You can't figure out who is profitable and who isn't, or how high up the curve you need to be to be profitable. Taking an average over all of the apps in the app store is like taking the average revenue of every website that tries to make money on the internet. It's a meaningless number.
At the end he compares the low numbers in averages to a single anecdotal example of something that had a 100 Million downloads. On the one hand he dismisses the few highly successful appstore apps but then uses another singular example of a highly successful app to recommend a different platform. Overall the methodological approach is deeply flawed.
I could go on, but ultimately, I found this analysis not very insightful. As with any new channel, it takes some time to figure out what works and what doesn't. Who is going to buy and who isn't. In almost all software the distribution is exponential where the top 10 to 20% make over 90% of the revenue. The same is true of the app store.
It all comes back to fundamentals. Find an unmet or under-served need in the market, figure out how big that market is, figure out how much it's going to cost to fill that need, make sure your app fills the need, and start trying to sell it. Then once you've built up a sales platform, then you start scaling it.
The author spends much effort estimating an 'average' app income and costs. Talking about averages does not make any sense in case of power law distribution. He is using Gaussian approach which is a wrong model for given phenomena. Bringing skewness/kurtosis/whatever to the table doesn't help a thing since he has completely mis-classified the problem.
> That said, independent developers making money on that app store directly is a complete crap-shoot.
Can the same can be said about entrepreneurs in general? What is an 'average' income of an entrepreneur (including failed ones)? Is it worth it? Do you aim for this dismal income when you start a business?
Scalable business model is completely different animal from a non-scalable one (i.e. selling your personal time as a developer, consultant etc.). Quite a few developers expect from iPhone apps the same they are used to in their contracting/salaried careers.
He doesn't. He tries to estimate the median precisely because he doesn't actually know the distribution although in several places he specifically mentions that it's likely not 'normal'. He's going for the median because it's more informative than the average, given the uncertainty about the distribution. It's a little moot, anyway, since the average figures are pretty lousy and given the most likely distribution, the median just makes it worse.
I just got an iPad and started using the App store for the first time in my life and guess what? Out of the 5 things I wanted to use the pad for, 3 of them didn't exist at all. There is plenty of space in this market I would say. Just not for iFart anymore.
My guess is that most branches of entertainment are similar. Movies, TV, music etc. There's no such thing as a steady, guaranteed income when it comes to things like this (in my experience/opinion).
That doesn't mean making apps is not a viable business. Just not for everyone who tries.
People write apps for iPhones because they dream to do what they love without working for the man. Of course it is only a remote possibility, but it is real.
99% of the people that went to California searching gold didn't find it, as in SouthAfrica, but the now deeper old mine was discovered by a solitary man looking for gold armed just with a mining pan.
There is something in the human spirit that is as valuable as money that makes musicians painters, singers,movie makers or engineers that maybe won't get rich keep trying. Maybe they get to work on what they love.
PS: I'm an iPhone App developer myself, since Linux Click an run app store I believed that a "right" implementation of an Appstore was going to be the future. I expected to be Apple the first to to this. In contrast with this executive they just seem to "get it".
Steve Jobs is really smart in this too, when he creates something he diminish it, so the competitor's executives see an appstore and say: It's only a mere millions dollar market, not important.
IBMs mainframes were the "important stuff", when real money was, mini computers were just "cheap" toys with no real application. People who wanted "job security" deciced to work with the "big boys" because it was safe... until it was not.
iPhone vs. WAP is a poor comparison because an iPhone app and a WAP site are two different digital tactics that address completely different objectives.
Brands are not buying iPhone apps to get the same utility that is available from a WAP site. They are buying them to make their brands appear technically savvy, and to increase engagement.
Can anyone justify an investment in development of an iPad-only productivity software? Do you think a startup can raise capital based on such economics?
My takeaway: Selling AppStore apps/ads isn't THE business model. It should be one of the channels amongst others.
Or, how does the picture look for games? Especially now that game developers can embrace a virtual currency / goods business model. They can get all the distribution benefits of being a free app while still making (potentially much more) money over a longer period of time. Or, do these apps not count as apps because they're too close to the web-based games that are doing so well on Facebook?
If you're a developer, looking at industry-wide numbers is good, but it's better to research the apps in the niche you plan on entering.
And, he's right. Big companies do take $50k+ to create mobile apps. Big companies do have much better things to do with their money; they can add far more value at greater scale. Of course, Jane developer might not be in the same boat.
Another interesting confounding variable: he clearly conceives of developers as 'one-app' developers. And, he's right that this doesn't pay on the apple store. There is just, frankly, some randomness in which apps tip on the store -- most developers have figured this out already, and of course Apple gives links so that developers have an easier time cross promoting.
More to the point, I think if you read the rest of his blog, and take the perspective of some of the larger international companies, going for gold with the iPhone won't be the best business strategy.
Tomi has given some case studies of how MMS has worked for car dealers in Germany, and even campaign promoters in Australia. There is this under-exploited niche called mobile marketing, which relies on WAP, SMS, MMS, etc which could give pretty high ROI for large companies (eg: BMW)
Especially if you take a B2B perspective, what he says makes huge sense.
http://communities-dominate.blogs.com/brands/2010/06/smartph...
What?
In any space with great competition and where your product is scalable, few will make enough money to justify their investment.
Also many paid apps are part of a bigger picture, the iPhone app is not the only revenue stream they have.
That's reality, now deal.
Just because the expected value of building something is low doesn't mean it's a bad model. It just means you have to execute well and that most people fail.
These numbers were really taken out of the ass of a troll.
A similar analysis could be done of PC shareware software authors, with similar results.
> So the best case of how few apps get to divide that 1.43 Billion dollars over the past 2 years, is 164,000 actual apps
You do have a point with in app purchases, but I doubt that they would significantly skew the numbers - they would make them worse for developers of paid apps, in fact. Given the other optimistic decisions the article has when dealing with numbers, I think that ignoring in app purchases fits with it: it's a best case scenario for paid apps.
It would be quite interesting to see how much money has been made through in app purchases, though. Not merely anecdotes, but the percentage of free apps which have in app purchases, and the average/median income through in app purchases in those apps.
He also fully acknowledges that there are some extreme examples that make it big, but that half of all paid apps are currently making less than $680/year.
This is a pretty thorough analysis.