What do users love about your app? What do they hate? Why do some hang around, and others leave? You can get all the answers when you do UX analytics!
In the introductory article to this series about app analytics, we established that you can analyze three broad areas:
- User experience and performance (UX)
Here, we will look into the first of them – UX.
Before we dive in, let’s just re-cap the basics. This will probably insult your intelligence, but here goes.
In order to retrieve user analytics from your app, you will need to integrate the SDK supplied by a specialist provider. This can take only a matter of minutes.
The SDK will collect user data and send it back to the provider to be processed and converted into insights. You will see these insights displayed in numbers and visuals on a dashboard.
The value of UX analytics is that they reveal how individual live customers are actually using your app (as opposed to tracking numbers of downloads, source of downloads, attribution, revenue, and so on).
So, most UX SDKs let you answer questions such as:
- Which parts of your app are users looking at most or least?
- Which areas of individual screens are users looking at most or least?
- What is their first gesture when arriving at a new screen?
- How long do users dwell on a specific screen?
- When do they pinch? Zoom? Tap?
- Where are they getting lost or stuck?
And needless to say, you can slice and dice the results. You can see different results by handset type or OS version, geography, demographics, and more.
This is valuable data. It helps you to resolve real-time performance issues and crashes.
For example, let’s say UX analytics reveal that Motorola phone users’ session times are much shorter than those of all other handsets. With this insight, you can investigate whether there is a problem in the way you have optimized your app for Moto devices.
UX analytics also help you to make long term improvements. You can discover and address important issues such as:
- Where users are leaving, and why.
- How to optimize the customer journey to reduce exits.
- Ways to rethink visual design to make the app more usable.
- How to tailor content to fit your users’ intent.
If you are very rigorous (and you have the funds) you can answer these questions pre-release. You can test with focus groups before you launch your product on Google Play. Simply get a bunch of people to put the product through its paces, sit back, and examine the results.
The alternative is to wait and perform on-going UX analytics with live users post-release.
In terms of scrutinizing the data, you have a range of ‘data visualization’ choices. The usual tools include bar and line graphs, journey maps, and downloadable CSV files. Heat maps are an increasingly popular visualization tool. With this option, you will be able to see an image of every screen in your app overlaid with a heat map that reflects where most of the user activity takes place. You can filter by UX type (tap, zoom, pinch, etc).
Perhaps even more useful is the ‘session replay’. This is a tool that actually recreates a user session in the form of a video so you can see the steps that led to an action (such as a crash, a session exit, a purchase, and so on).
Meanwhile, many providers now offer recommendations on best practice actions you can take to improve your app. They base this on their aggregated observations of what works best across the industry.
There are many many UX analytics providers to choose from. They all offer a free service with tiered pricing based on the number of events you track and the range of tools you want to access.
There’s a decent guide to other products here.
It’s a little overwhelming. So Google Firebase is probably a good place to start (though it does much more than just UX analytics).
With a good UX analytics tool in place, you are well set to solving performance issues and finessing your app to give users the best possible experience.
But what about downloads and installs? How can you assess where users are coming from, and how long it takes them to on-board?
Well, that’s the realm of engagement analytics. Next time we’ll look more closely at this topic. Check back soon.