Heap SQL: Path Analysis
This query allows you to monitor the most common flows through your product. As is, the SQL selects the first 5 events of every session consisting of at least five events. It then aggregates the number of users who have completed each flow and displays the top 20 most common paths. Looking at the table below we see the most common user flow is
signup_enter_password and that this particular flow has occured 2791 times.
Heap is unique in the way it captures events due to the flexibility offered in defining events. A given
event_id across all Redshift tables corresponds with a unique event recorded by Heap. That said, the same
event_id may exist in multiple event tables because Heap provides the flexibility to create multiple event definitions that correspond to the same raw event. For instance, you may define the following two events in the product:
Click CTAdefined as
Click on .cta=
Click CTA - Homepagedefined as
Click on .ctawith a
filter where Path equals /
If a user clicks the CTA on the homepage, a new event will be recorded on both event tables because it corresponds with both event definitions. As a result, two events with the same event_id will be included in the all_events table as it contains every recorded instance of all defined and custom events.
In order to select the events most valuable to you, when multiple events are firing at the same time, this query selects the events that are the most specific. This is based off of the heuristic that events with mores specificity are fired less over all than events with more general definitions. If you look at the example mentioned above, although two events are firing when a user clicks
.cta on the homepage, this query would only include the
Click CTA - Homepage event, rather than the global
Click CTA event, allowing you to have more insight into your users paths.
In Mode's visulaization, each block of color is representative of a particular event; it's size relative to the percentage of users who have completed that event. By hovering over a particular chain of events starting at the center of the circle you can trace the relative and absolute percentage of users who have followed the path within your app.