Retention: The Revenue Catalyst
While you might expect a 10% increase in user retention to lead to a 10% revenue rise, the reality is far more complex.
I’ve talked to hundreds of intelligent product managers, and each of them knows that retention is crucial for the app's success. At the same time, most fail to answer how their revenue will change if retention increases. To truly understand the interplay between retention and revenue, we need to dig deeper into the metrics and use models that provide a panoramic view of user behavior.
Apps vary significantly in their monetization methods. But like we usually do in physics, let’s create a model to help us understand the complex reality. For simplicity's sake, imagine the app Sticky as a B2C product where users pay $1 at the end of every week they're active. No usage, no payment. This assumption is handy because retention directly transforms into revenue. Different business models like subscriptions, freemium, or one-off transactions are more complicated to calculate, but the approach is the same.
When 10% is 10%
There are a lot of ways to understand and measure retention. I've written a separate essay comparing various methods. I recommend doing that before proceeding further if you haven't read it yet.
In making an informed decision, we need to look at two distinct retention types: one for new users and another for active ones. No one-size-fits-all indicator captures both, so we'll dive into a key metric for each type, which best reflects the impact on revenue. N-week retention focuses on new users and is closely linked to cohort analysis. On the other hand, the Active User Retention Rate is all about active users, no matter when they started.
With N-week retention, calculations are simple, and our intuition holds true. A 10% increase in user retention leads to a 10% revenue rise. Let’s set up a small model for our hypothetical Sticky app to double-check the logic. We had 10,000 installs in week 0, and after 12 weeks, we retained 284 users from the initial cohort. This gives us the Retention Curve, with the corresponding revenue being $11,865. We'll label this as Variant A, shown in dark grey on Chart 1.
Now, imagine we enhanced the onboarding experience, which bumped our 4-week retention up by 10%, moving it from 9.4% to 10.4%. This becomes our Variant B, colored orange on Chart 1. The increase in revenue is also 10%, amounting to $13,051 for Variant B. For a more detailed look at the model, please follow the link to the spreadsheet.
10% increase in retention = 1.5x more revenue
When it comes to the Active User Retention Rate (AURR), straightforward math no longer applies. The catch is that we're dealing with calculations similar to compound interest. If you’ve ever tried to understand how much you will get if you put $100 on a deposit for 10 years at 3% interest, you know you need to be good at math to crack the task. Interest is calculated not just on the principal amount of money but also on the accumulated interest of previous periods. This means that the amount of interest grows exponentially over time if the money is left to compound.
There are different ways to calculate AURR. For example, Duolingo defines it as, "The chance a user comes back this week if they came to the product each of the past two weeks." To simplify the model, let’s assume that Sticky's Active Users are those who returned this week and used the app in the previous week. In this edge case, our AURR will be equal to Week-over-week retention.
When you make improvements for the active user cohort, you increase retention throughout the whole customer journey. And if you do this, you increase AURR for every week. It’s like a snowball rolling downhill: the bigger it gets, the faster it grows. When we keep more users active for week 1, we have a better starting point for week 2, which influences week 3.
Imagine that we improved messaging in Sticky, so our AURR grew by 10% from 89% to 97%. This is represented as variant C, which is orange on Chart 2. The earnings uplift is a notable 50% (!), with the revenue for variant C standing at $17,850.
10% AURR may be hard to achieve in the short term, but even more modest growth leads to a significant revenue increase. Retention is a true revenue catalyst.
In reality, unlike the model, a 10% retention increase will give you more than 1.5x in revenue, especially when considering a timeframe beyond just 12 weeks. After achieving a 10% retention growth, your business will undergo a substantial and positive transformation.
What AURR is good?
As we can see from the model, the Active User Retention Rate has a compound nature, making its relation with revenue nonlinear. But what is the interplay? What AURR can you consider a good one?
To answer these questions, we'll plot another graph. The X-axis is the AURR value and the Y-axis is the number of users who churned after 12 weeks. To simplify, let's assume we have no new active users.
There are quite a few graphs and numbers in this essay, but Сhart 3 is the most important one. To me, it is very dramatic. If your AURR is 60%, the business is critically ill. For prosperity, you usually need at least 80% AURR. Relatively small changes in the latter part of the chart can be transformative: a change from 90% to 95% almost doubles the number of users.
All calculations are in the spreadsheet.
Retention is all you need
Retention and its growth are all you need to get your company prosperous. Uri Levine, the Waze founder, stated in his book that retention is the only metric to measure product market fit achievement. If the retention curve flattens, you’ve got PMF.
I’m convinced that constant focus on retention improvement is what differentiates a mediocre business from a true market leader. Sure, you can focus on attracting new users instead of retaining existing ones or (OMG!) reactivating dormants. In the short term, this approach can yield positive results. But if you play the long game, regular personalized engagement will gradually increase the number of lifelong customers willing to spend more often with your app than newcomers.
Our on-the-napkin model was a toy example. But the real-world charts look very similar. If you improve your retention today, grateful users will bring you more money next month, the month after, and for years to come.