Windows Phone 7 developers got an unexpected present from Microsoft this holiday season: early payouts (bumped up to January from February) and immediate access to downloads and sales data (originally slated for February as well). I don't know why Microsoft had a sudden change of heart, but I do know that there was a widely circulated and quoted, copied, and imitated article about how hard it was to make development decisions regarding the server-side end of Windows Phone 7 apps without this data. The timing of the change in course for Microsoft and the publication of that article (November 29, 2010) is curious indeed.
Before this information was released, developers were really frustrated because they knew the data existed. Bing's Visual Search was showing the "download popularity" of apps for a short period of time (incidentally, I think that the ranking weights recent downloads much more than past downloads), and developers were finding other ways to get the data (such as pinging a Web service when an app was run for the first time or purchased).
A look at the numbers
From my perspective, the numbers are positively abysmal. I am not going to comment on how these numbers may indicate potential handset sales because I don't know nearly enough about the correlation between the number of handsets sold and the actual sales numbers on other platforms to make any guesses. I have said that Windows Phone 7 development should be thought of as a hobby or a learning experience rather than a business model, and I was spot on. Let's look at the data for my Windows Phone 7 apps to see what I mean.
My first Windows Phone 7 app was the free Airport Status Checker. It's hardly a unique app, and it's limited in scope. Incidentally, the app has a poor user rating (one-and-a-half stars from three reviewers, last I checked), but since a good number of the downloaders are outside the United States, and it only works for U.S. airports, I suspect that is the source of the poor ratings. At one point, Bing Visual Search showed it as #272 in the rankings, but it has been slowly slipping. A friend reports that his apps are slowly slipping, as is my other app that has been out a while, which is why I believe that recent downloads are more important in the rankings. All the same, it is still at the bottom of the top 10% of applications on the platform, which isn't too shabby.
Airport Status Checker had just over 900 downloads in a month, but let's put this into perspective. Every now and then, I post source code that accompanies one of my TechRepublic articles to my company's Web site, and those packages get over 900 downloads in several days, so 900 downloads does not impress me. That's the number of downloads that some obscure, niche piece of shareware can get on one of the major shareware/freeware/trialware sites. Of course, the phone market is much, much smaller. At the same time, the Windows Phone 7 App Hub is too. It isn't the saturated market that the Apple App Store and the Android Marketplace are. You can write a throwaway app like Airport Status Checker, and it will still stand out a little (it didn't help that I used a public domain icon that a number of other apps used too).
But what do the numbers on a paid application look like? My Local Crime Rate application has been ranked at the top of the bottom 50% for a while, coming out around #1,700 and sliding down to around #2,000. It has been out nearly a month. It is priced at 99 cents, and does not have a free trial. It has a whopping seven downloads. Assuming the pace of sales holds steady, accounting for Microsoft taking 30% of revenues and holding checks until you've earned $200 in payouts, it would take me over 288 months to see my first $200 check from Local Crime Rate. And that is an app that is not doing too poorly in the overall rankings. That tells me that apps in the bottom 50% are looking at 10 downloads per month, or less.
My third application, Name That Nerd, provides some additional clues. After about a week in the market, it is rated #2,006 (out of 3,000 listed). It has 24 downloads and 1 sale. Extrapolating a little bit more, that's 100 downloads a month and 4 sales (at 99 cents each). I can understand the low sales numbers -- it's a quiz application, and I simply did not put many entries into the quiz database (I was in a hurry to get it approved so I could enter a contest). I take full credit for the sales numbers.
The top apps are pretty high-quality, for the most part. Kudos to the folks who put forth the time and effort to making something good. (There are reports that Microsoft has been slipping payments to developers to port apps or develop apps for Windows Phone 7.) That said, my Airport Status Checker app shows that you can be in the top 10% (which isn't a bad place to be) and not be moving many units. If it was a paid app, even with a 20% conversion rate (which is phenomenal), that would be under $200 in sales because Microsoft took its percentage. While $200 in sales is a good number for an application that required this level of effort, it is abysmal for anything that I would have spent much time and effort on.
My advice still stands for the time being: do not develop applications for Windows Phone 7 with the expectation that you will make big bucks. Does this mean that there is no opportunity in the Windows Phone 7 App Hub to make money? Not at all. I think that the market is wide open for apps.
Games and entertainment dominate the top apps, but outside the top tier, the apps look pretty run-of-the mill, with a lot of unit converters, flashlights, and dice rolling applications. The applications are mostly uninteresting right now. I think that the folks who can fulfill basic, mass market need (say, an outstanding Twitter or Facebook application for 99 cents), put together a polished game, or create a must have business application will be on to something. I am sure that there is a logarithmic rise in download rates in the top 100 as well.
So, if you have an app that you think can get into the top 100, or it will be something with staying power and earn enough per month to make you happy, by all means, go for it. But if your plan is to count on the sheer number of downloads to carry a niche or less than stellar product into profitability, you are mistaken. Windows Phone 7 just does not have the number of handsets out there for a "throw mud to the wall and see what sticks" strategy to get a great success unless you have a way of cranking out mud at a rapid pace.
J.JaDisclosure of Justin's industry affiliations: Justin James has a contract with Spiceworks to write product buying guides; he has a contract with OpenAmplify, which is owned by Hapax, to write a series of blogs, tutorials, and articles; and he has a contract with OutSystems to write articles, sample code, etc.
Justin James is the Lead Architect for Conigent.