Thursday 24 November 2016

An Intro to Mobile App Analytics

business2community.com
Colorful bands leading up and away from the viewer

Make better decisions with data–that’s the promise of analytics. Web analytics are already an established part of most marketing and product teams–tracking pageviews, click-through rates, and other metrics. But what if your organization relies on mobile apps? User behavior on apps is so different from the traditional web experience that it requires a very different way to approach analytics. In this article we’ll look briefly at what mobile app analytics are, how they’re different from web analytics, and some of the major mobile analytics tools.

Getting Started: Ask the Right Questions

A good way to think of analytics is as a way to answer questions about your users and customers. In order to make the most of whatever data you gather, you first need specific, well-formulated questions. Are you trying to figure out what people do inside your app? That’s engagement. Are you trying to understand how to keep users around longer? That’s retention. Are you trying to understand who exactly your users are to begin with? That’s user segmentation.
What questions you ask will depend on your business needs and goals. Those questions, in turn, will help you identify the kinds of data that are most relevant to you. For example, a retail app will likely want to know how much people are browsing their different offerings and whether there are places where people abandon the checkout process. By contrast, a video streaming app may be much more interested in knowing how much time people spend watching different videos and finding ways to improve their watching experience. Meanwhile, the makers of a freemium game app will likely want to know how best to entice their users to buy upgrades, expansions, and other paid content. Each of these scenarios involves a different set of metrics and different approaches.

The Challenge of Mobile: Beyond Pageviews

How are mobile analytics different from web analytics? The answer is that they’re as different as the experience surfing the web on a desktop and using an app and the business models that support them. On the desktop web, users are used to switching between several (or more than several) tabs and scrolling across complex and fast-loading web pages. Here, metrics like pageviews, monthly unique visitors, and ad clickthroughs are traditionally the most important metrics. But these are often less meaningful for users of mobile apps, which tend to be both simpler in design and more immersive than websites. That’s why mobile analytics focus much more on metrics like session length and retention.

Thinking About Events and Funnels

Once you’ve settled on what questions to ask, how do you collect the data that will help you answer them? Some analytics tools (Google Analytics, for example) will automatically gather certain kinds of data, including the number of users, the duration of their sessions, the device types they’re using, and where they’re located. While this general data is certainly valuable, in order to get more specific insights about your users and how they’re interacting with your app, you’re likely going to want to implement some form of event tracking.
An event is simply a defined user interaction. This could be playing a video, “liking” a friend’s post, bookmarking an article, or just about anything else. Each analytics platform will have a different means of defining and recording events, but nearly all of them will require some amount of coding from your mobile engineering team. These events will then be logged and saved in a database, ready for analysis by data or marketing teams.
Series of events can also be strung together into flows or funnels. These can tell you a more complete story of how a user is engaging with your app. For instance, if you want to identify where customers are abandoning their cart in your retail app’s checkout process, you could put together a series of events beginning with visiting a product page, adding it to the cart, proceeding to the checkout screen, entering credit card info, and finally tapping “buy.” A well-constructed flow would allow you to see exactly where customers are dropping out, and allow you to specifically target those parts of your app that most need improvement.

Choosing an Analytics Tool

Selecting the right analytics tool depends on your specific business needs. Some analytics tools are specifically geared toward developers: These might emphasize features like reporting crashes and exceptions, real-time reporting, A/B tests, proactive alerts, Integrated Developer Environments, and API integration with bug tracking and task management tools.
Other tools really emphasize the value for the marketing team. These tools place a greater emphasis on features like installation tracking, cohort analysis, measuring CLM, marketing automation, and mobile ad attribution. Here are a few of the most popular mobile analytics platforms on the market, as well as what sets them apart.
Flurry has long been an industry leader in mobile analytics, used by tech behemoths like Snapchat and Skype. Part of what makes Flurry so appealing is its ease of integration: developers can get analytics up and running in a matter of minutes, compared to hours or days with some other platforms.
Mixpanel is another industry leader, used by a number of tech behemoths including Uber, Airbnb, WordPress, and others. It has a robust set of features focusing on both engagement and retention, as well as real-time reporting which makes it great for teams that need to iterate quickly. Even though many of its features are geared toward developers, Mixpanel also includes a point-and-click interface that allows non-technical folks the ability to ask questions without writing SQL (or bugging the data team). That said, integration with Mixpanel is more involved than with some of its competitors, and it can get pricy as operations scale up.
Localytics is aimed more at marketers in large organizations. As such, marketing-critical features like installation tracking, CRM management, and marketing automation are available out-of-the-box.
Amplitude has some of the most advanced user behavior analytics suites, which makes it popular with marketers. In addition to tracking demographic info like location, install source, and platform, Amplitude also includes a uniquely robust funnel analysis tool as well as allowing you to group users based on their actual actions rather than demographic data.
AppSee places a lot of emphasis on understanding user interactions in more ways than just conventional event tracking. To that end they offer user recordings and touch heatmaps that allow developers to see exactly how users are engaging with an app.
Remember: Data is only valuable if you can capture and measure it. Whatever analytics platform you go with, you’ll want qualified data engineers to implement your tracking and interpret the data once it rolls in. Depending on your business needs, you may need marketing and product resources to formulate questions and set objectives, UX and UI design those new flows and features, and engineers to implement them. Building a successful mobile analytics operation is a cross-functional undertaking.

No comments:

Post a Comment