The user journey: disengagement and reactivation
Part 4 of 4: User journey guide for Google Play apps and games
This article is part of a series of articles by the Google Play Growth Consulting team that focuses on increasing app performance by optimizing the user journey. (You can also catch up on our previous topic, understanding KPIs for apps and games, in this article and webinar series.)
We’ve worked our way through each of the early stages of the user journey: onboarding, activation, and commitment. Now let’s look at the final stages of a user’s journey — specifically, what happens when users become disengaged and how to reactivate them.
This article will cover questions like the following:
- How should I define lapsed users compared to churned users?
- What are some normal disengagement trends across different types of apps?
- What is a re-onboarding flow and why does it matter?
Let’s dive in.
Disengagement, in context
It’s completely normal for some users to go through periods of disengagement; in other words, it doesn’t necessarily mean your product design is unsuccessful or ineffective just because a user lapses. Here are some examples:
- A user may be consciously reducing screen-time.
- If your app is seasonal, a user’s interest in specific content may be decreasing during the off-season.
- A user may be less in need of what your app provides (usage of a dating app may go down as a user’s relationship status changes).
However, you do want to understand which users are starting to disengage — and more importantly, why. Disengaged users often provide important clues for optimizing your product and serve as the proverbial ‘canary in the coal mine,’ helping to identify potential issues in your design that could be contributing to lower retention.
Disengagement, defined
Disengagement comes in two forms: short-term (lapsing) and long-term (churning). Every app is different, so how you measure disengagement should vary based on your app’s ideal usage, user expectations, and other considerations.
We recommend defining lapse and churn in 1-week increments. Comparisons between cohorts are easier when you use the same length of time. For example, if a user has not had a session in the last 7–13 days, we call them 7-day lapsed users. If a user has not had a session in the last 14–20 days, we call them 14-day lapsed users.
We reserve the term churned for users who haven’t shown up in 28 days or more. Differentiating between lapsed and churned users will help you identify unique behavioral patterns associated with short-term and long-term disengagement.
App categories and disengagement, at a glance
For many apps, lapsed and churned users can be a significant portion of your audience and therefore represent a large opportunity to grow your business if you can bring them back. Below is what the disengaged population looks like across a few categories from top-performing apps on Google Play.
In the dating app category, 42% of users lapsed at least once for 7–13 days in a month, 26% lapsed for 14–20 days, and 12% lapsed for 21–27 days. This is a significant number of users. Seeing these trends, it’s clear that some disengagement behavior will be normal for your category, while other disengagement behavior may indicate there’s something in your app design that requires your attention.
Understanding the various motivations behind disengagement can help draw a clearer picture of where to focus next.
Motivations for disengagement
We’ve touched on how user engagement can fluctuate due to things like seasonality, evolving user needs, or digital habits. Let’s get more specific.
There’s a difference between a ‘voluntary’ lapse (when a user knowingly decides to use an app less often) and an ‘involuntary’ lapse (when a user either unknowingly ignores an app or can’t use the app due to reasons out of their immediate control).
Think of voluntary lapse as lapsing with intent. Users in this category have likely disengaged because of changing circumstances, which cause them to make a deliberate decision to use the app less. This type of lapse is difficult to predict or mitigate, as it often happens without much warning and could be caused by external factors. One example of this could be storage space on a device (this is particularly relevant for apps with a global business, specifically in regions where average devices are cheaper and tend to have less storage). Take a moment and think about the last time you chose to use an app less or delete it entirely. Why did you stop?
Involuntary lapsing, on the other hand, is essentially lapsing without intent. Something urgent or unexpected could have come up in a user’s life or lifestyle. The user could have simply forgotten or been distracted by something. With these types of scenarios, you can (sometimes) predict and prepare for them. Well timed marketing campaigns through email or push notifications can help remind users about the service. For example, a food delivery app sends push notifications right before lunch or dinner. Take a moment to reflect on an app you actively used but haven’t used in a while. Why did you start using it less without realizing it?
For users who leave your app involuntarily, disengagement can be a useful tool to predict when a user will likely lapse. That’s because declining engagement can be a great predictor of involuntary lapse.
A reliable metric to measure this trend is to look at the time between core actions (for more on what a core action is and how to use it, see our introduction to the user journey). An increase in time between core actions indicates declining engagement.
For example, say we’re working on a popular fitness app. Because we track core actions, we know our median committed user completes anywhere from 3–5 core actions (logged workouts) in a week. Now let’s say we notice two new trends emerging: over the past few weeks, the core actions per week for our median committed user has declined to 2.5, and the time between each core action has increased. If this trend continues, user engagement will decline further, leading to lapse or churn.
Quick case study: Clue
Clue, a period-tracking and health app, occasionally asks their users to update their goals directly in the app. This helps the team know if a user is voluntarily lapsing or if they are forgetting to use the app and involuntarily lapsing.
Clue noticed their user base had lower return rates than they expected, and after they sent out short in-app surveys and reviewed the results, they realized there were a few use cases that they weren’t actively taking into consideration in their data analyses.
For example, users who were pregnant no longer needed active period tracking on a monthly basis. This is an example of when lapse behavior doesn’t necessarily indicate that a user lapsed or churned from the app because of dissatisfaction or product-market fit (two reasons that, under normal circumstances, would prompt a team to reconsider its feature roadmap or messaging strategy).
The Clue team then decided to introduce an ability for users to temporarily pause push notifications and reminders. The benefit of this feature was that the team now had a better signal about those users who temporarily lapsed and allowed them to better personalize their reactivation efforts.
This is an example of what’s called a ‘pulse check.’ Deploying (thoughtfully crafted) in-app surveys can help you understand how user motivations may be changing in ways you aren’t yet thinking about. It’s also a helpful tool to identify use cases that lead to disengagement, and, ultimately, can help you tailor reactivation campaigns and win them back.
Considerations on reactivation
Understanding how to effectively reactivate users is just as important as onboarding them and keeping them committed. As mentioned, this can represent a large portion of your potential user base.
It’s helpful to consider the user’s state before they churned. Some examples include whether or not the churned user was a recent install, whether they were committed (according to your own definition) before they left, and how much time they spent in each stage).
- For instance, let’s say User A is committed and User B is a new user. If both users involuntarily churned (28+ days since last login), it would be difficult to reactivate them. However, it would still be easier to bring back User A than to bring back User B. This is because User A has already demonstrated a product-market fit, and something could have happened that made them involuntarily churn. Perhaps surfacing a recent feature update or content release could reactivate User A, while User B may not have been the right product-market fit for your app to begin with.
- Now, let’s say that User C and User D were both committed, but User C has recently lapsed whereas User D churned a long time ago. Which user is easier to reactivate? Well, it depends on the reason why they left. Assuming both users churned for similar reasons, we would have a higher chance reactivating User C compared to User D. That’s because User C has only recently lapsed, and we know that it’s considerably easier to reactivate a recently lapsed user than a long time churned one.
The image below shows the relative effort of re-acquiring users who have lapsed and churned. Generally, it’s easier (and cheaper) to reactivate users if they’ve been away for a short while. This is especially true for engaged users (activated and committed) who have developed a habit engaging in core actions at a set cadence. The longer they’re away from their routine, the harder it is to reactivate and re-engage them.
The key is to understand who is disengaging and when so that you can craft the appropriate reactivation strategy for each segment of the population; early intervention targeted at the right segment of users can catch users before they disengage further and inevitably churn.
A note on re-onboarding
Lastly, don’t forget that if you reactivate someone, you’ll also need to onboard them again.
All our best practices on onboarding still apply. Here are a few tips specifically on re-onboarding users who have lapsed or churned:
- Consider having “welcome back” messaging to acknowledge the user’s status and make them feel like they’re receiving a more relevant or personalized experience from your app. You can also have “pick up where you left off” messaging, if your app retained the user state.
- Depending on how long a user was gone, your app may have new features to introduce or could have been redesigned in ways the user may not be familiar with. A few “what’s new” pointers and callouts are a good way to catch users up on the latest improvements.
- At a minimum, you want to reintroduce returning users to the core action. This ensures that they re-engage with the app and once again return to a committed state.
- Bear in mind that users might be experiencing re-onboarding numerous times so allow them to skip the re-onboarding should they choose to.
In our experience, we find that many apps don’t offer re-onboarding flows for returning users. This is a huge missed opportunity. We recommend creating a returning user flow and, additionally, making a user’s re-onboarding experience as smooth and enjoyable as the first time.
A good practice is to consider re-onboarding while you’re designing onboarding flows. Ask yourself if the information you present to a user when she opens the app for the first time is relevant for when she comes back after lapsing or churning. For each onboarding screen, ask yourself if a lapsed or churned user would benefit from seeing this information again. If so, how might you tweak the information to be more appropriate? If not, exclude it from the re-onboarding flow entirely.
Conclusion
If you consider the end-to-end user journey and how your user base is moving through and behaving in each stage, it will be much easier to create product strategies and roadmaps for your app that reflect the biggest opportunities.
As we’ve explored in this four-part series, every user has his or her own expectations and motivations. The more you understand these cohorts, trends, and patterns, the more you can set your business up for sustainable, long-term success with a target audience that is the right fit for what your app has to offer.
Remember, it’s engagement that drives your business. Graduate your new installs to become onboarded and activated. Habitualize activated users to turn them into committed users. Re-activate and re-onboard disengaged users through tailored re-activation campaigns and re-onboarding flows. The key is to pay close attention to your users — and business success will follow.
If you missed the other articles in this series:
What do you think?
What metrics are you most or least familiar with when analyzing app performance? Let us know in the comments below, or tweet using #AskPlayDev and we’ll reply from @GooglePlayDev, where we regularly share news and tips on how to be successful on Google Play.
