Retention Curves as the most important metric — a better alternative to MAUs
In a previous blog post I wrote why I don’t believe in Monthly Active Users (MAUs) and why, especially when measuring Product Market Fit (PMF), your most important metric should be customer-focused and specific to your product.
But I was recently struck by the perspective that there’s another great indicator to measure PMF that doesn’t require you to define product-specific metrics: retention curves.
It’s surely a less charming metric than MAUs, but it’s an equally transversal one (doesn’t need to be product-specific), and which gives you a great picture of your customers’ interest in the product over time.
We can break retention curves in 3 parameters:
Initial drop-off rate
The initial drop-off rate represents the % of users who have signup but never showed back after the first day.
For example, it’s common for users to install and signup an app on their phone but never get back due to being buried in the big amount of other installed apps, or for not being that relevant for their use case, etc.
As in many other metrics and parameters, this varies from product category to product category.
Due to high initial drop-off rate in some cases, you may notice this section to be hidden in some retention curves: instead of the curve starting at 100% and day 0, they start at 70% on day 1.
Rate of descent
Then we have the rate of descent, which, well, is quite self-explanatory, to be honest.
You can use this to compare different cohorts (eg: based on age groups) and evaluate which have a higher rate of descent — which cohort “loses interest” faster.
Terminal Value
The terminal value answers the question:
What % of customers who tried the product continue to use it in the long run?
If the curve goes to zero, then it means the product eventually loses all its users.
This is super valuable information. For example, in many products, you will not be pouring money into the start of the funnel (as marketing) to know that soon enough all your users disappear. Some things need to be fixed sooner.
Nonetheless, none of the parameters is more important to the other I would say: analysing the 3 of them can give you different insights about where the product is failing and where to improve. This can then lead you to further investigation through the use of qualitative research as User Interviews.
Maybe you’ll not even need to analyse each one of these parameters individually (I haven’t) — the important point is that, by looking at retention curves we can have a strong indicator of the level of PMF a product has, and how it evolves over time:
Products with higher PMF will have a lower initial drop-off, lower descent rates, and higher terminal values. Products that are improving PMF over time should result in the improvement of retention curves as well.
Is that simple, but very strong analysis.
Next time someone tells how well a product is doing, ask them for their retention rates. They will give you a better picture than MAUs and many of other vanity metrics.
I’m Daniel, Product Manager at Whitesmith. Paper Planes is a place where I reflect on my experiences and learnings on the craft of Product Management, and where I share them with my team and community under the form of short blog posts.
If you enjoy these, please show your love by tapping the clap button!
You’re also welcome to follow me on Medium or Twitter!
