In the wee hours of the morning, around 2:00 I think, I released our first Android application.
(Google's Marketplace doesn't really work except with their application, and their application is not available in all countries, nor via all carriers, nor on your PC -- a topic for another post. You'll get an error, otherwise. You will soon be able read some about it at http://sfsmart.com, but our site is still under construction. My marketing partner is unexpectedly travelling this week, so we don't have as coordinated a launch as we would like.)
It's been an interesting experience getting here. I began learning the Android platform at the start of the year, using this application as a learning platform before we undertake more serious efforts. It's been a better choice for learning than I could have imagined.
Over the next few days, I'll post some of the things I've learned, and try to give some insight into what it's like to pick up new skills and enter a new market. I hope to cover both some of the more technical aspects that help in creating a solid product, and share some of my thoughts on marketing, from a developer's perspective.
One of the decisions we made was to go with the Free/PAID model, where we have two applications, a Free version, supported by advertising, and a Pro version, which dispenses with the advertising and offers some additional features. Today's release was the Free version.
The launch appears to have been a success. When I woke up this morning, and checked the stats, we had something like 122 downloads in 6 hours, from all over the Northern Hemisphere. Later in the day, we got our first comment in the Market -- quite positive, with a 5-star rating (out of 5). Our little trial application appears to have nicely solved a real problem for a real user -- that feels good!
One of the decisions I made was to include Flurry monitoring into the application. Without being intrusive in any way, nor disclosing any personal information, it gives me insight into how people use our applications, an any potential problems.
Indeed, among the mornings statistics were several occurrences of one error, which was initially concerning. Further investigation showed, however, that these were errors that I'm handling and reporting, and should not be impacting users. This is a problem I found during some of my extensive stress-testing of the application. I left this temporary workaround in place to find out how often it would happen in practice. Rarely -- only a handful of sessions out of thousands, but still worth fixing.
One of the frustrations of the Google Marketplace is that people leave complaints there that you have no way of helping them with. One of the first things I did after downloading our app was to leave a message requesting people send support requests and problems to our email address, but of course, I haven't gotten any email at all (except for test messages I sent to verify it works!)
But already one user has left a semi-complaint that one feature didn't work for her. I'd like to better understand, but I can't reply and ask for more information without wiping out my other message. I'm hoping I can identify the problem through the Flurry data, but not so far.
It may be a documentation problem, or a limitation of her device, or even a bug. It's going to take some more proactive digging to find out, when a simple email would have been so helpful, and not much more trouble than a comment.
Still, we don't have a right to expect users to go out of their way to help us. But we really DO appreciate it when they do. Even in this form, if I can identify this woman's issue and resolve it, she'll have been helpful. But please -- we'd really like to see just positive comments on the market, and get all questions and problems sent via email. It would make us look so good! :)
In the meantime, I've been monitoring our ad revenue. We're up to $0.70 which doesn't sound like much, but I'm pretty happy with that so far. How good that is depends on how much people continue to use our app. If we just serve ads to people who try it out and then discard, it won't be so good. $1/day will pay some of our expenses while we work on our next product, but won't be enough to build a company on.
On the other hand, if we end up with a growing set of old users, we may see that number climb.
I don't believe it's possible to make a living on ads for a single utility like this. I think you need to get a number of applications out there, and a mix of free and paid.
Of course, there are the big hit applications which download tens, of not hundreds of copies. These are in a different class. That's a great kind of success to have, but it's also a very unsure kind of success.