Josh · May 20, 2013 at 06:00 pm

App Analytics, or the Death of the Independent App Developer

This is entry number one in a series of articles about tools for app developers, today focusing on app analytics. The goal of these articles is to spur conversation in our new app development forums, so go there and contribute your experiences.

I was on a plane last week, sitting next to a 15-year-old Japanese girl. I was about to pull out my GS3 to play some CCS (Candy Crush Saga for the uninitiated) when I looked to my right and noticed that she was doing the same. A tad embarrassed that I was playing the same game as a girl with a pink Hello Kitty phone case, I got up and walked the aisles. On a not-full 737, I counted six other people actively playing the game. No one was playing anything else.

If you’ve ever played CCS, you know that it was designed from the ground-up to be social and viral. It integrates with Facebook so well that you will find yourself passionately competing with people you haven’t spoken to in a decade. It hits you up for in-app purchases at the exact moment of addiction, but still gives users who would never pay a dime a path to win (and share with others who might be more likely to pay). The actual game is a mediocre knock-off of Bejeweled, which itself is probably a mediocre knock-off of ten prior products. Once you acknowledge that, you realize that you’re obsessed with a game in which far more effort was put into “forcing” you to keep playing (and maybe paying) than innovating with game mechanics. And this is the product dominating the mobile charts.

For this article, we interviewed several providers of app analytics tools: Apsalar, App Annie, Appboy, and Flurry. These are all valuable and powerful tools, and each product has a different focus, each company a unique business model. We also spoke with a developer (who wished to remain anonymous) at a major Silicon Valley startup who insists that Google Anaytics, probably the “default” tool, especially for web devs turned app devs, is all he needs. The conversations we had focused around terms like “viral coefficient,” “user invite process,” “social distribution,” “customer segmentation,” “behavioral segments,” “MAUs,” “DAUs,” etc.

I completely understand if all you’re thinking is, “ain’t nobody got time for that!” There are over 1.5 million apps in the various mobile app stores, and while there are still a few remaining categories dominated by individual developers (including Android tools where XDA members shine), that pool is shrinking. And this is where the XDA app development community diverges: If what you care about is making a cool, free product for the community, we salute you. And truly, those people should stop reading, because you probably don’t have time for any of this. For those who dream of millions of downloads and making a living from apps, ignore analytics at your peril.

Cezary Pietrzak at Appboy tells us, “Building a great product is table stakes, but it’s only a start.” He goes on to give some advanced tips, telling developers to focus on community, content, and context.

  • Community: “The smartest apps ask only their most active users to rate them or share them with friends, because they expect a much higher response rate among this group.”
  • Content: “Using push notifications or in-app messages to serve micro-content (rather than plain alerts) can drive engagement significantly. Email is also very effective…”
  • Context: “The best apps use location data, behavioral triggers, historical usage patterns and other data to create a very relevant, contextual, and personalized experience. Fab alerts you about new sales, Foursquare tells when your friends are nearby, and Circa sends notifications about stories you follow.”

So what are the most valuable pieces of data you can pull from an app analytics package and how can you use that data? Christian Poppelreiter, an account specialist at Flurry says, “Flurry Analytics has sections which report on session duration, session frequency and overall rate of retention as an application ages. Developers can customize how they collect data through events tracking…such as when someone likes a status, shares an article, beats a level, or makes a purchase. Once events are set up, developers can also segment out sections of their audience according to behavior (i.e. purchasers) or according to more traditional audience metrics like age, gender or location.

“There typical length of a session can tell you how many ads might be appropriate to place in an ad supported app. Developers can track how long users typically spend within different sections of the app, can detect when users are most engaged and also see conversion rates.”

The point of this article isn’t to recommend one analytics tool over another, and we’re reluctant to link to company marketing collateral here. But in this case, we think it’s worthwhile if only to emphasize to our typically-independent developer crowd how serious the business of analytics has become: Apsalar pointed us to a case study with, in which they were able to “double their viral coefficient” by leveraging “our funnel analysis to identify a key bottleneck in their user invite process.”

If all of the above sounds daunting, we agree. In fact, we can imagine that it’s rather depressing to most readers: The Zynga/Farmville-ization of apps *is* sad. 95% of users won’t scroll past app number 50 in Play, and it feels like the only way to get there is to release something derivative and then bug/coerce/manipulate users into sharing.  Sure, thanks to intense competition among tool providers and the free-ness of Google Analytics, most analytics and marketing tools are easily accessible to any developer, even free in most cases; but the reality is it takes significant time and resources to make the tools work for you. This is not a part-time pursuit for a team of one. In many ways, it feels like the death of the independent app developer.

But we truly don’t want to discourage. XDA is about community and unique ideas that drive the future of mobile development. Maybe our users aren’t building the next CCS, but we’re innovating. Call us idealists, but we think great ideas can still get noticed. Maybe they won’t make you Zynga rich, but they will get downloads and usage—especially among the early adopters here. If that’s the approach you want to take, then here’s the best advice we got during our conversations: Christian at Flurry says, “Focus on your customer experience and design the app the way that you would want to use it if you were the customer. Many developers start with a great idea, but compromise the user experience with something that is either poorly organized, with limited functionality or something that is overrun with advertising. People download apps because they want to perform some kind of task, whether that task is sending a message, reading an article or playing a game. They don’t want to feel disrupted, nor do they want to feel like they’re being given a hard sell, and I think a lot of developers need to tread carefully on the fine line between what engages the user and what earns them money.” We couldn’t agree more.

All of the companies discussed in this article offer some sort of free analytics solution, so it can’t hurt to check them out. If you’re still confused about which one is the best for you, check out our full interviews, which we’ve posted to the forums.

How have you used analytics tools to increase downloads and improve engagement in your products? Do you think the indie app developer stands a chance? Is building a “great” app enough or should the indie developer just give up? Knowledge-sharing, discussion, community are key to keeping independent app development alive. Please share your experiences in the forums.

Want something on the XDA Portal? Send us a tip!


js1999 is an editor on XDA-Developers, the largest community for Android users. View js1999's posts and articles here.
Emil Kako · Apr 26, 2015 at 04:19 pm · 3 comments

Which Lockscreen Security Type Do You Use?

From pattern locks to the controversial face unlock, there are a number of different ways you can secure your Android phone's lockscreen. Some methods are clearly more secure than others, but it comes down to user preference at the end of the day. So, which lockscreen security type do you prefer and why?

Chris Gilliam · Apr 26, 2015 at 12:00 pm · 4 comments

XDA Recap: This Week In Android (Apr 18 – 25)

Here in the digital XDA newsroom, we spend our days pouring over an average of 2,500 news items and forum threads every 24 hours. Only the most timely and interesting bits survive the editing process, but the portal's front page still sees weekly counts in excess of 100 posts. This is a glut of content to absorb, especially if following the news cycle isn't your full-time job. However, the tech world is vast, and the information must flow. With this in mind, please...

Mario Tomás Serrafero · Apr 26, 2015 at 11:00 am · 3 comments

Sunday Debate: Custom ROMs vs. Modular Tweaks

Join us in a fun Sunday Debate on Mods and ROMs. Come with your opinions and feel free to read some of our thoughts, then pick your side or play devil’s advocate to get your voice heard and engage in friendly discussion. You can read our food-for-thought or jump straight into the fray below!       PLEASE NOTE THIS DEBATE IS ABOUT FUNCTIONALITY AND PRACTICALITY AND NOT ANY IMPLICATIONS. LET’S AIM FOR FRIENDLY DISCUSSION Tweaking Android is what a...

Share This