4 Things to Look For In Mobile App Analytics

shutterstock_202180546Understanding how your mobile apps are performing sounds like a simple thing. When you gain users and corresponding revenue, then you’ve been successful. But while Gartner predicts that the mobile app stores will generate $77 billion in revenue by 2017, not everyone will share in that bounty. According to a study by VisionMobile, more than half of all mobile developers bring in less than $500 per month per app.

Mobile performance can be the factor that separates the winners from the also-rans. Negative reviews in app stores can torpedo a mobile project’s chances to get downloaded. And app performance (freezes, crashes, and slow speeds) is one of the biggest lightning rods for user complaints on the app stores.

Smart developers can use mobile analytics to see what is causing apps issues so they are able to fix them faster and spend more time developing new innovations instead of addressing festering problems. That can lead to better perceptions of the app’s brand and ultimately a better return on investment when building mobile apps.

Here are four key areas to pay attention to in mobile analytics:

1. Code performance

It isn’t necessarily that your code is bad… it may just be that it isn’t performing exactly as you expected. Code-level analytics data helps you see exactly where the problem is in your mobile app—and how to fix it.

The first step is the ability to look over the shoulder of your users and see exactly what they see as they navigate your app. But you need more than that. To provide a better user experience, with New Relic you can also learn not only how long each page took to load, but also where that time was spent.

Interaction Traces help you see what is happening on the back end when users are using the app—were delays due to database calls, memory usage, CPU usage, network requests, or other reasons.

Finally, tracking and making sense of the fire hose of performance analytics can be overwhelming. So developers look for tools that not only collect and return performance results but also help filter and visualize those results to quickly identify exactly what needs to be changed.

2. Device peculiarities

Because the mobile ecosystem is so fragmented, testing every iteration of every mobile operating system on every device in a reasonable time frame is pretty much impossible. According to a study from Flurry, to reach 90% of active users requires supporting a whopping 331 device models!

That matters, because mobile apps really do perform differently on different devices—and mobile developers need the details. It’s helpful to know that your Android version earned rave reviews for responsiveness even as your iOS installations are being savaged by unhappy users. But that information isn’t actionable unless you know exactly which versions of the OS—running on which devices and which carriers—are causing the problems.

You should be able to know within minutes of a new product release if it’s causing any issues with your apps. More importantly, you need to be able to identify those issues and the specific steps to remedy them so you don’t alienate previously happy users as they update their devices.

3. Network performance

Mobile networks are not created equal, and their performance can have significant effects on how your apps perform. Knowing whether a hiccup in your app is caused by your code, the OS, or the network can make a huge difference in how you approach remedying the problem. To quickly isolate problems, you should be able to filter slowdowns, HTTP errors, and IP errors by carrier and region.

While delays and other problems due to network issues may not be your fault, you still should be aware of them and figure out ways to deal with them so they don’t alienate your users.

4. End-to-end visibility

Because so many different things can affect mobile app performance, it’s important to go broad as well as deep. Mobile developers can benefit from having a single place to see the big picture of everything that’s happening to their apps. Of course, seeing individual events at the device level can be key, but knowing what’s going through the network and what’s happening with the backend Web services that are powering the app is also important.

Putting all those elements in the same place can help developers focus on solving the right problems, understanding the dependencies among them, and establishing clear responsibility for issues. Some things may require action from the mobile team, for example, while others may involve the Web, and it’s important to know who has to fix what first.

See this eBook for more information: Slow Mobile Apps Ruin Everything: How to Drive Growth with Better Mobile Performance.

Smartphone image courtesy of Shutterstock.com.

Scott Koegler is a technology journalist with 20+ years experience writing about enterprise topics, cloud computing, application development, integration, supply chain, big data, and interviews. View posts by .

Interested in writing for New Relic Blog? Send us a pitch!