May 24, 2013

Don’t Iterate in the Dark, Use Analytics to Drive App Development

Mobile app analytics packages like Flurry provide invaluable insight into who is using your app and how they are using it.

Posted on January 22, 2013 by in Process

Iteration is the beating heart of the app development process. Of all the apps we’ve built, I can honestly say that none have come out looking or behaving like the app we envisioned when we began. Apps have this funny tendency to evolve and twist their way into something else by the time they find their groove. Building an app is less a straight-line path from idea to app, but more like a cycle of design, build, release and measure.  To stoke the fires of iteration, we keep our release cycles short (3-6 weeks) and we use app usage data we gather from each release to drive what we do in the next cycle.

But how do we gather app usage data?

Well we could try sending comment cards and survey monkeys to people, but I wouldn’t hold my breathe on that. No my friends, to get insight we use mobile analytics to instrument every one of our apps so we can track exactly who is using our app, how they are using it and what they are doing with it.

For those of you who come from the web world, you probably have heard of Google Analytics. Google Analytics gives you incredible insight into the audience of your web site via traffic data gathered by Google.  Mobile analytics suites like Flurry and Localytics to provide the app equivalent of what Google Analytics does for the web sites.

You wouldn’t launch a web site without connecting it to Google Analytics, then why would you release an app that didn’t plug into some instrumentation suite?

We use Flurry Analytics, and with that we get a free web-based dashboard that gives us deep insight into how people use our apps. Just by adding a very small amount of code we immediately gain insight into user retention rates, user growth rate, session lengths, user demographics, and a whole bunch of other usage data. With this data we get visibility into our user base and what pieces of functionality are resonating with them. This allows us to tinker, tweak or totally revamp our app’s in between releases with real data to support it.

Here is a brief snapshot of some of the app usage data we can track through our Flurry web dashboard:

Flurry provides session and new user data for our apps

User growth and usage statistics like these give us insight into how much usage our app gets over any time horizon.

flurry provides insight into the devices and operating system of your users

Device and operating system statistics of users of our app.

Flurry gives us some idea of the type of app user that is using our app.

Flurry gives us some idea of the type of app user that is using our app.

App usage broken down by geography

App usage broken down by geography

Session length distribution tells us how engaged users are with the app

Session length distribution tells us how engaged users are with the app.

I rarely speak in absolutes, but when it comes to mobile analytics and instrumentation there is no excuse for not incorporating something like Flurry into your app. It takes about 3 minutes to modify your app to work with something like Flurry and it’s free! Without instrumentation, the only thing data points you have to measure your app’s public reception is the number of downloads and app reviews it gets, which as Will Smith so eloquently stated is “just fishing in the dark, son.”

Read more about integrating Flurry into your app here. (No, this isn’t a sponsored post, I just woke up today and realized I really like Flurry)

Bobby Gill (64 Posts)

Bobby Gill is Editor-in-Chief of IdeaToAppster.com, author of "Appsters: A Beginner's Guide to App Entrepreneurship" and the founder of Blue Label Labs, a mobile development lab based in New York and Seattle. In addition to building its own apps, Blue Label Labs provides design and engineering services to mobile app clients across the world. Prior to starting Blue Label Labs, Bobby spent 4 years at Microsoft as a Program Manager within the Forefront Identity Manager (FIM) product group. During this time, Bobby served as an engineering and architectural lead for the FIM server specializing in database and web service design. After Microsoft, Bobby worked as a Summer Associate at McKinsey & Co. where he was part of a service operations enhancement program within the high-tech industry. Bobby holds a Bachelor of Mathematics specializing in Computer Science from the University of Waterloo and a Masters in Business Administration from Columbia Business School.


Tags: , , ,