Three weeks ago, we decided that we are going to focus all our efforts for our app Sporzer towards increasing the number of Monthly Active Users (MAUs). The goal we set is to bring this metric up at least 10% per month. We can do that with new signups, or getting more of or users to activate and come back.
What is Sporzer?
Sporzer is a fitness and bodybuilding social app we're building for iOS, Android, and the web. You can take a look at sporzer.com.
In the process of creating a little dashboard to keep track of the Monthly Active Users, it got me thinking: what actually is a monthly active user?
I Googled around a little bit and came across a few metrics how to articles. All of those seem to calculate a user who does something once in a month towards the Monthly Active Users.
I think this is unfair and leans towards a vanity metric - a metric that might look well but doesn't say shit about the success of your company. Treating monthly active users in this way would onlywork if you can lock in the primary use case or revenue conversion on a single activation per month, and this is not the case for most apps.
Improved Monthly Active Users definition
So how to improve? What is a better definition for Monthly Active Users for Sporzer?
Sporzer is a social app where there are primary features like posting your (daily) fitness pictures, keeping track of your goals and showing other people how good you look. That means that for someone to get value from Sporzer, they need to use it on a regular basis.
So I decided to go with an improved Monthly Active Users definition. One that also includes a retention aspect:
A monthly active user is someone that uses Sporzer at least twice a month, with at least seven days
between usage moments.
It took some mucking about with custom queries to get this out of Mixpanel, but eventually I was able to generate the following dashboard with some custom querying.
To illustrate, here's our metrics dashboard without the improved monthly active users metric:
And here it is with taking into account that an active user should come back a second time in a month with at least seven days apart from the initial event:
You can see in the previous two screenshots that the initial non-improved Monthly Active Users graph shows much higher numbers than the second graph. These figures are higher because they include people that only did something once in that month. The second graph is a much more realistic view of the Monthly Active Users on Sporzer since it only includes people that did something twice, with at least seven days in between.
Don't use a vanity metric as your goal for higher level business decisions or prioritization. Think about what an activation means in your product and when an activation is good enough to say something positive about your customer paying you actual money or getting real value from your app.
I think adding a retention component to the Monthly Active User metric is a great way of filtering out the vanity part for Sporzer. You might need a different filter for your app!
Questions or comments?