Experience of an Android App Launch

This is my long overdue post on the launch of Droid Weight last Thursday. I had been looking forward to publishing my first app very much because I was incredibly curious what would happen.

Before continuing with that story however, I would like to take the time to thank my awesome friendly users (aka testers). Thanks so much for helping me avoid the worst of launch troubles: an app full of minor annoyances and major bugs! Your help was very much appreciated.

In a way the first app was a learning experience and an experiment. I’ve been asked “So what happend? How many downloads do you get? Are you adding more features now?”

Let me try and answer all these questions.

I’ll start with a short chronology of events, the ‘what happened’ part.

Timeline

  1. Step One: Publish. I published the app – which took far longer than expected because you have to provide quite a bit of information for the market about your app. I like the way how this is very much a guided experience so you can’t mess up very much. (Templates, languages, descriptions & sizes for graphics etc.)
  2. Step Two: Wait. Immediately I searched for the app and I was badly disappointed: it didn’t show up. Somehow I got used to everything happening at the push of a button. The market took it’s sweet time. After a couple of hours I could finally find the app when searching for “droid weight”.
  3. Step Three: Disappointment: I could not create any other search query in which my app would show up in the first 10 pages of results – I never bothered to look farther back, users don’t. So I prepared for total failure.
  4. Step Four: Checking the Developer Console. (simply finding it was difficult enough). Maybe I was overlooking an obvious query and there were downloads. Nope. None.
  5. Step Five: Reloading. Checking the Developer Console again. Around midnight I got some fresh data and there finally were downloads! I was relieved – not a complete failure then.  It turns out that the Developer Console only gets new data every other day, so I wasted a lot of time on Friday by reloading and hoping for more numbers.

So these are the download numbers for a random free app in a rather busy market segment: there are more weight tracking and diet apps out there than good metaphors.

Droid Weight does not even show up in search queries so how do these numbers happen? I found out soon enough: at least on one search query it does show up on the first page and when I found out I was quite happy that I had taken the – very little – pain to translate everything to German which is after all my first language. If you search for “gewicht” (German for ‘weight’), then Droid Weight turns up at #5 (actually #7 but free/paid versions I count as one app).

So much for my first experience of publishing an android app to the market. Here are my conclusions:

Lessons learned – stuff I did right

  1. Localization: I’ll try and add at least French to my repertoire for localization. It is easy and it is totally worth the time.
  2. Presentation: I took a day to prepare the launch on the market. I created all the required graphics and all the optional ones like a promo graphic or a banner. These are the first things the user sees about your app. The app icon is of course part of this as well. I will repeat the effort for my next app.
  3. Friendly User Test: mine was incredibly helpful. I got so many reports and a few feature requests that I managed to not go live with any major bugs and I would have probably!

Lessons learned – stuff I botched

  1. Key Words: when preparing the presentation for the market think of search key words you want your app to show up for. Do not forget your key words in your presentation! I really messed up on this one. I knew about it and I was in too much of a hurry to write up better marketing texts. I should have taken time to carefully prepare all the texts with search queries in mind ahead of time.
  2. Development Process: despite best intentions and previous experience my development process turned into blind stumbling quickly. This ended up costing me a lot both in terms of time and quality. I think part of my failure comes from using an excel sheet to track this stuff 🙁 Bad tooling in other words. I should have made a design, a (project) plan and an estimate how long it would take me to finish. I should also reconsider my tooling.
  3. Unique Feature: I wanted to have one but had to let it go because everything took longer than I thought it would (see #2). Next time my unique feature gets a higher priority and I think I’ll try harder to find something special to implement. Rolling out just “more of the same” is not going to cut it long-term.

There’s probably more. Those 6 are the first I thought of. All in all for a first effort with a random idea I am happy: after all I finished and launched a project with a new technology within a month of first starting to use that technology!

More features, need more features …

Right now I am busy adding some features and doing some refactoring. Here’s the feature backlog:

  • redesign handling of preferences in the app
  • add a minimum size for the graph and scrollbars for smaller displays
  • add body fat tracking & graphing
  • re-add body measurements tracking (waist circumference etc.) & graphing
  • adding a home screen widget to enter your weight
  • add a configurable daily notification that reminds you to track your weight
  • adding goal entering as either target BMI or target weight

So which one seems the most promising? What should I do first? Any other suggestions? I have budgeted 1 more week until I want to start on the next project.

Update: added last week’s numbers to the graph. Two weeks of downloads and I am up to about 700 downloads and 400 installs. That’s kind of nice.

This entry was posted in apps. Bookmark the permalink.

Comments are closed.