The way customers use apps is fundamentally different than how they interact with your organization across other channels. As a product owner, you know this to your core, and have created an app with that in mind - one designed to meet the needs of mobile customers and ensure that their experience is rich with features that engage.
With analytics tracking in place, you can identify common traits within your user base, how they interact with features and navigate the app, and what retention looks like. But sometimes, these metrics and reports are too static to give you a deep understanding of your users.
To get to know your customers at an even deeper level, you need to be able to leverage predictive analytics.
Predictive insights reveal those rare moments where users discover core product value, and are highly indicative of future behavior. For product managers, what this means is a better informed and more efficient roadmap for delivering core product value to your users as quickly as possible. To learn more about exactly how this can be done, read on for insights and examples.
Changepoints, or the top related user behaviors, can lead you to discovering where the “Aha!” moment is in your product. Essentially, what part of the product is of the most value to the end user? Take Facebook for example: users who add at least seven friends in their first 10 days as a user are much more likely to become repeat users. Here, the top related user behavior is the addition of a minimum number of friends within a specific time period. So, the question then becomes: how can Facebook influence new users to add friends and hit this benchmark in order to improve retention?
If you can identify the earliest time in the user lifecycle in which customers are delivered core product value, you can work to change the onboarding process, optimize the UX or UI, or update potentially substandard components of the experience in order to prompt users to arrive at that “Aha!” moment faster and more efficiently. Knowing this will also help you uncover churn risks within the app - namely, what are the top user behaviors related not just to retention, but also to churn?
A publishing app uses predictive analytics to discover that one of the top related user behaviors to churn was users firing a ‘Low Memory Warning’ event on Android. This uncovers a previously unnoticed suboptimal part of their Android product experience, in which users were frequently getting low memory warnings. They are then able to assess the cost of this suboptimal experience and make a more informed decision about the resources needed to make this improvement.
Roadmaps are a complex combination of needs driven by a host of influencers - and we understand just how much energy it takes to maintain a balanced one. Which is why app data and predictive analytics, in particular, can be hugely helpful in driving prioritization, proving need and optimizing your roadmap. Primarily, you can use predictions to understand big product investments that are needed to customize and improve the app experience for different segments of users. This often means looking at the common user attributes (vs. behaviors) that are most related to churn. Not only does this give you deeper insight into your customers and their profile attributes (age, gender, location, etc.), but it also surfaces correlations between those attributes and possible churn risks.
Global brands are often faced with the challenge of localization. Take, for example, a retail app that sells and ships across the globe but only has one app version, built with a US audience in mind. In looking at their user attributes, they uncover that users in Asia are more likely to churn because their experience isn’t localized. Therefore, the decision to create an app version specifically for this market became a lot easier when they saw that the top related user attribute to churn was User Country = China.
They might have also seen that users who are on older devices, operating systems or app versions are more likely to churn and encouraged them to update their OS or app version. Or, it could be a reason for the team to make an investment in improving the product experience for older devices.
Tracking retention across your app user base is key to understanding the stickiness of the experience, particularly when you introduce new features, app versions, or UI/UX updates. What you need to know is how these changes are affecting retention in the short- and long-term. Supplementing this data with predictive analytics gives you a more robust database of insights.
Specifically, you can use the predictions themselves to understand how rolling churn rates are changing over time, as opposed to traditional retention graphs which show how cohorts change over time and the change in between cohorts. By tracking with predictions, you have the ability to test product changes and observe churn rates before and after the change was made, in addition to changes in retention.
One way to use this data is to see how changes over time occur on a seasonal basis and then design a product roadmap which coincides with seasonal changes. For a Sports app, many users become at-risk of churn leading up to the end of a particular season, for example, the end of football season. This could be an indication that they should schedule some feature work to release to these users at that time of the year. They can then track the changes in churn rates before the feature release and after, and compare that to historical season-end retention data.
Thanks for signing up. Look for your first email shortly!
We’ll reach out shortly to schedule a time to talk.