Funnels let you analyze a collection of events, such as a user sharing their first article or successfully changing options on a settings screen.
Funnels provide a mechanism for analyzing key processes within an application. Most apps have a clear goal such as getting a user to make a purchase or upgrade to a paid subscription; these often involve multiple events. Use a funnel to analyze any collection of events.
Funnels are retroactive and respect the same filters and date ranges as other screens so it is possible to experiment with many different funnels and see changes over time. Funnels provide very granular data. More than just counts and filters, use filters to explore successful and unsuccessful conversions at every step to identify what is causing success vs failure.
See the guide on Analyzing Funnels for an in-depth exploration.
-
Select any step to see the performance of that step by comparing it to the
previous step.
You cannot see details of the first step because there's nothing to compare it to.
The detailed results display including the conversion rate by day as well as a list of successful users (in blue) and unsuccessful users (in red) below the graph.
- The blue graph on the left shows the actions that users who are successful tend to take before converting on this step.
- The red graph on the right shows users what dropped off performed after the previous step instead of this one. This helps identify what users are doing in place of making it through the funnel.
-
Scroll down to the last chart to explore these users in terms of any Localytics
dimensions, where there is an interactive split of the red and blue users.
You can:
- Split by devices to discover that the experience is much worse on a slower device or one with a smaller screen.
- Use a funnel tracking registrations to discover that users who start the process by signing in via email quit more frequently than users who sign in via a social network.
- Have a funnel tracking article reading to sharing the article with friends and using attributes to find what sections are most likely to cause a user to share.
- Split by acquisition campaigns to see if certain campaigns get users who are better at converting than others.
- Use country and language splits to see if region or localization is creating a barrier to success.
