In February 2012 I decided to learn how to make mobile apps. I owned an Android phone and I happened to have known the Java programming language in which Android code is written. So for me choosing to learn to code on Android was just a naive coincidence.

I didn’t have any unique app ideas at that time so for my first app I recreated the functionality of my website at that time which was a site to openly discuss problems and create businesses to fix those problems. The website was called Problemio.  By April 2012 I released the first version of my app.  It was this business plan app called Plan & Start Small Business.

Just to paint the picture of how bad the first version of the app was: it didn’t even have a logo. Instead of a logo, it had the default little green Android man.  The app crashed on many devices because I didn’t include an error reporting mechanism and wasn’t even aware when it crashed and for whom. Sounds terrible, right? Of course! But Android let me get away with it and partially that is the beauty of Android and the freedom of creativity it allowed me to have.

I just kept working on the app as I wanted to.  I had one thing working in my favor. Unlike in Apple’s system where developers have to wait a week or two for their app update to go live, Android allowed me to update the app as often as I wanted.  And I took full advantage of that, at times making up to three smal app updates in a single day. I would wake up, observe analytics from the day before, make and release changes according to what I observed. By afternoon, my server logs would give me some sample data of what was happening on the app and how my morning changes affected things. Then I would make an update based on that, and again look at the logs in the evening evening, make more changes, and go to sleep to see the results of the evening update when I would wake up the next morning.

That cycle of innovation allowed me to quickly improve the app and to my surprise, by the middle of summer, based on 10o or so user reviews the app had at that time, the app was the highest rated business app on Android with a 4.8/5.0 rating. Since then, I started to try to monetize the app and the rating dipped a bit to 4.7/5.0 which still makes it one of the top rated apps.

So what is the secret of such a highly rated app? The app is not impressive at first glace.  The design is not as good as it should be, and parts of the usability still needs to be improved.  But having made over 150 mostly small app updates, I was able to test every variation of the main usability flow in incredible detail. I tested how users would behave differently for every order of buttons on the home screen and got to understand how people are using the app at a very deep level. And from the deep understanding of how people were using my app, I was able to make the app very helpful for my target users.

So the questions is: what would my experience have been like if I had started on the iOS platform instead? I can answer that because sometime during the middle of June 2012 I began learning iOS and finished my first iOS app that August.  It was mostly a clone of my original Android app because I wanted to be on both platforms. For those curious, here is that iOS app. It is named Business Plan & Start  first of all, because I made the app in a rush while learning iOS development, just as the Android app once was, you can guess that the iOS app was pretty bad at first.  Even though I submitted the app to Apple in August, they rejected the first version of the app, and I was only able to have the app go live in the App Store by mid September. In contrast, during my first few weeks live on Android, I was able to fix most of the very bad mistakes. But instead, on iOS, I had to fix the one thing their review team found wrong, and wait two weeks for them to review it and approve the app before I was able to continue to iterate. That meant that the two weeks was almost all wasted when it came to the ability for me to iterate on the app.

And that iOS cycle went on. Considering the two week period for an app update to go live, after six months, I was only able to make a maximum of 12 updates (I actually made fewer) which is nowhere near what is needed if you are innovating and looking for a product-market fit. In fact, the only way I was able to get around that is to continue iterating on my Android app, and only putting features into the iOS app which I was able to thoroughly test with many frequent Android releases. I continue to work with all my business apps the same way.