Events for Flurry Analytics¶
Custom Events are Flurry’s way of tracking specific in-app actions your users take. Examples include playing a song, taking a photo, completing a level, and so on. This page walks through how to interpret the Event data for your application.
Note: For developers looking to test that they have tagged Events properly, please review the Event Logs section below. This is the first place that Event data populates in the Flurry portal, typically within a few seconds.
The Event Summary view provides a tabular view of all Events from the collection of apps selected in the Context Blocks. The icons in the table allow you link directly to the Event Details page or the Parameter Distribution Page.
Event Details Dashboard¶
For every Event that you have tagged in your app, Flurry provides a few different basic metrics. To view these metrics, you can either click on the Event Details icon of the desired Event from the “Event Summary” found at Metrics > Events > Event Summary or go directly to Event Details view by clicking it in the left nav.
The Event Details dashboard gives you a number of different views into the Event data recorded by your app. Please note, as with any dashboard in Flurry, you can click on the chart to view it in full-screen mode where you can also download the data behind the chart into a CSV.
The following provides a review of each of the charts on the dashboard.
Events per Session Distribution¶
Provides information on the distribution of the number of times an Event is triggered per session. This is helpful to understand the frequency of the event in your users’ sessions – for example, this chat app might want to know how many times per session the “sendChatText” Event is triggered to get a sense for communication volume. This chart would inform the team that users typically send 1-2 texts per session and in less common circumstance send 3-9 texts.
Unique Devices Performing Event: The next chart displays the number of unique devices that have triggered the Event in a given time period. Click the cog on the chart and use the Time Grain controls to view this on a daily, weekly, or monthly basis. Please note that because this is a unique view of devices taking an action, each device can only appear once for a given time period. For example, if a device A sends 30 texts in July and device B sends 2 texts, they are each counted only once in the July tally. Here we see that in approximately 5.3K unique devices sent a chat text at least once on March 11 2017.
Displays the total number of Events that have been triggered over time. Again, click the cog on the chart and use the Time Grain controls to view this on a hourly, daily, weekly, or monthly basis. Continuing the earlier example of “sendChatText” event, this would tell you the total number of times that texts were sent in a given time period.
Average Event Duration¶
For any event that records duration, this chart displays the average duration over time. This is essentially the total time spent doing the event during the period divided by the number of events that occurred in that period.
Event Duration per Device¶
For any event that records duration, this chart displays the average duration over time per device that triggered the event. This is essentially the total time spent doing the event during the period divided by the number of devices that triggered the event in that period.
Event Duration per Session¶
For any event that records duration, this chart displays the average duration over time per session. This is essentially the total time spent doing the event during the period divided by the number of sessions that occurred in that period. Please note, this metric can show a declining number in recent periods in the case where a meaningful number of sessions are ending when the user is offline or the data cannot be sent. If you see this, please reach out to firstname.lastname@example.org as this can be indicative of an integration issue and in that case we can help.
Event Occurrence per Device: Displays the number of Events that occur per the number of Devices that trigged that event in the period. This is essentially the total count of Event Occurrence divided by the number of devices that triggered the event in that period.
Event Parameter Reporting¶
If you have added parameters to your Event, you can view this data by selecting the Parameter Distribution icon in the Event Summary table. You can read more about Parameter Distribution here..
In addition to the Parameter Distribution view, there is a Parameter Trends view that allows you to review a timeseries of the occurrences for selected values across up to five different Event Parameters.
The Events that are recorded by your app can be used to investigate the path users take through the app. This functionality is available in the Explorer section of Flurry. You can read more about this feature here.
Event Logs provide a detailed view of all of your users sessions that contain Events. This is the first place that session data hits the Flurry portal; all other charts subsequently update with this information. Data populates in this section within minutes of the actual session.
For developers testing their Flurry Event integration, this is the place you should look to see if your Event testing is working as expected.
The Event Log provides a log of the Events (and associated parameters ) that were triggered in a specific session. For each session it displays:
- Device type
- UserID (if the developer has passed this to Flurry). For more information, see Advanced Features for Android or iOS.
- Events and parameters triggered (Note: parameters viewed by clicking the ‘+’ next to the Event name)
Please note that if a session did not generate any Events, it is not displayed in the Event Logs.
Was this document helpful?