Using Funnel Analysis to Measure User Conversion Rates


Mobile Analytics Best Practices

In previous posts, we discussed using cohort analysis to measure the retention, engagement and monetization of your users.

Also important is having the ability to measure user conversion rates on goals.  As an app publisher, you want app users to take a set of actions toward desired goals.  In a game, this could be completing a level or purchasing virtual currency.  In a shopping app, this could be completing a purchase or sending a product recommendation to a friend.  These goals are specific to what you’re trying to achieve with your app. In each instance, you need the ability to measure how well users are converting on these goals in order to improve.

Funnel analyses are an effective way to calculate conversion rates on specific user behaviors.  Think of a funnel as a series of steps toward a goal.  In an app, these goals are important user actions of your choosing.  The goal you seek to measure could signify retention (relaunching the app), engagement (reaching level 10), or monetization (completed in-app purchase).  You may even want to see how well you are converting users on upgrading from your first app to your second app.  In each case, a funnel analysis is the report of choice.

When building a funnel, you have the ability to establish the goal event, as well as each step leading towards this goal.  These steps can be within your app, or across two apps.  Keep in mind that the steps aren’t a specific user path.  Instead, they are the steps you’ve determined a user will likely reach on the way towards a goal.  A simple goal could be a user completing an “In-App Purchase” for a mobile game with the following set of steps:

  1. Launch App
  2. Reach Level 2 of Game
  3. Offer Displayed
  4. Clicks on Purchase Offer
  5. In-App Purchase Complete

Once the funnel has been established, you will then begin to see exactly how many users are converting on these goals, giving you the insight to know which part of your app you will need to optimize to improve conversions.

Let’s look at three examples:

  • Completing in-app purchases – what percentage of users spent money on a particular digital good?
  • Level Progression – what percentage of users reached Level 6?
  • Cross-selling your other apps – what percentage of users in App 1 went on to download and launch App 2?

In-App Purchase:

We created the funnel below to assess how well users are converting for an in-app offer.  As you can see, of the 13,289 users that entered the funnel, 1,088 users completed the purchase, giving us a total conversion rate (Funnel CVR) of 8.19%.  This important knowledge is useful when comparing how well users were converting on this purchase in March versus February.  You also have the ability to see conversion rates between individual steps, which could highlight any steps that are producing significant drop-off.  If you do see a peculiar drop between steps, it’s an indication from your users that something in the user experience is creating lower conversion rates during that particular stage in the process.

Measure Conversion Rate Funnels

You can also segment users by their behaviors to see how different groups of users convert versus one another.  A segment could be as simple as “Returning Users” or could be much more complex like “Users who have spent more than $10 and are from Canada.”  In the example below, we took the same In-App Purchase funnel from above, but this time zeroed in on our first time users.  As you can see, the conversion rate dips to 6.37%.  This would represent an opportunity to improve the experience of this app’s first time users.  The developer or marketer could then decide to perhaps offer a tutorial or another mechanism to help with the introduction of the app.

Mobile Analytics Conversion Funnels

Level Progression

This funnel was created to understand how engaged users were in December by calculating how many reached Level 6, which represents a significant achievement in the game.  In this example, we see a total conversion rate of 11.52%.  Similar to the example above, we can also see the conversion rates between each step.  This will tell us if there is something wrong with a level – for instance, if not enough users are progressing to the next level.  Perhaps the level is too difficult, too long, or not engaging. Either way, it gives us concrete evidence that a change to the app is needed.

In this example, we see a substantial drop off between the first and second step.  It seems as though we have many users that never reach the first level.  This is something that clearly needs to be addressed with further analysis. Getting more people to reach that first level would greatly improve this app’s total funnel conversion rate.

Funnels measure and improve conversion rates

Cross-selling your other app:

In this example, we created a cross-app funnel with just one step to see how well the offer in App 1 was converting users to downloading and using App 2.  Nearly 5,000 users viewed the offer to download App 2, and roughly 1% of them actually converted.  This creates a great baseline to compare and improve upon over time.  With this information, you now have the ability to measure whether changes to your offers improve upsells or cross sells.  This is also a good opportunity to A|B test different offers in different locations of the app.

Mobile app analytics report

As you can see, funnels are an incredibly flexible way to measure user conversion rates on goals within your apps and provide important insight into how well your users are engaging, retaining, and monetizing. Simply put, funnels are a critical component of any complete analysis of user behavior and should be used by any publisher serious about improving their mobile business.

If you have any questions, feel free to drop me a line at or sign up for our free app analytics.

Share Button