Different count in Analytics, Segmentation, Campaigns stats & Flows

The difference in Campaign Counts

On the MoEngage dashboard, user behavior analysis can be analyzed on different pages - Analyze, Segmentation, and Campaign stats section.

Segmentation Counts

The Create Segments page always provides "unique users" count.

Analyze Counts

Analyze section pages - Behavior, Funnel, Retention, etc. pages provide two different kinds of counts

Total event count - This is the count of the action/event done by the user. The total event count is available for hourly, daily, weekly, and monthly granularity.
Unique user count - Unique user count is the count of distinctive/individual users performing the events, a user can perform a specific event multiple times. Unique user count is available for hourly, daily, weekly, and monthly granularity.

The following Analysis types: Total Event and Unique users counts can be seen in Behavior & Funnels. While Retention always provides a unique user count.

Campaign counts

Here we have taken an example of the Push campaign -

Impressions represent the number of user devices that have received the notification sent by the marketer. It can also be seen as the total event count of impressions.

Clicks represent the total number of notifications clicked by the users once the notification is delivered. It can also be seen as the total event count of Clicks.

Conversion Events represent the total number of primary conversion goal events that happened within the attribution period from the time the user received the notification.

Conversions

For campaigns having only one recurrence, conversion represents unique users who have accomplished the primary conversion goal event at least once within the attribution period.
For campaigns having multiple recurrences, conversion represents unique users who have accomplished the primary conversion goal event at least once within the attribution period for each recurrence. For example, If a user receives one Campaign Instance on Day X and another on Day Y and converts once on both days, the conversions will be counted as 2.

Difference in count

Event count for Impression, Clicks, and Conversion Event should be the same in both Campaign stats and Total Event analysis of the Behavior page.
As these metrics are event counts, they can not be compared with Unique Users.
Conversions in campaign stats provide the unique user count per campaign recurrence, and Behavior analytics Unique Users provide the unique user count across the hour, day, week, or month.
For example, a user who receives two campaigns and converts two times Monday & Wednesday (within the attribution window) - this user will have campaign conversion as 2 (one for each campaign). In contrast, in Behavior analytics, Unique Users for daily granularity provide 2 users and weekly granularity provides 1 user.
Campaign stats are calculated in UTC and all the analyses provided in the analyze section run on the app time zone.

Note: In Funnel analysis, Funnel - Unique User analysis with Holding attribute constant on campaign id attribute across all steps, provide unique converted user for each campaign for the selected granularity. Read more about it here.

Funnels and Flow Stats Counts

The disparities in the results of Funnel Unique Users Analysis and Flow Metrics stem from their different methodologies. While the Funnel displays unique user-level conversion counts, the Flow Metrics are calculated at the Unique Trips level.

As an example, if a user enters the Flow on two separate occasions, such as on the 1st and 5th of a month, and completes a conversion once in each trip, the Entry count and Converted Trips will register as 2 in both cases. Conversely, the Funnel analysis of Unique Users will reflect a count of 1 for this user. This discrepancy arises from the Funnel's emphasis on unique user activity, rather than the number of trips undertaken.

Upon comparing the results of Funnel - All Occurrences with the Flows Trip metrics, the count of User Entered Flows remains consistent.

Note: To simplify the querying process, we recommend using "User Entered Flow" and "Flow Trip Conversion" events. This ensures that only users who have entered and engaged with the flow are considered when generating the funnel view. You can use attributes such as "Flow ID" and "Version Name/ID" to filter for specific flow versions. To differentiate between user types, you can utilize the "With Control Group" attribute in these events.