Segments

A segment is a subset of your users. For example, of your entire set of users, one segment might be users who come from a same traffic source. Another segment might be users who trigger certain events, for example, made a purchase.

Access Segments

To access Segments, follow below steps:

1. Go to Clickstream Analytics on AWS Console, in the Navigation Bar, click on "Analytics Studio", a new tab will be opened in your browser.
2. In the Analytics Studio page, click the Segments icon in the left navigation panel.

Create a segment

1. Click on Create Segment + button
2. Input Segment name and Description
3. Specify a Refresh method
   • Manual - Segment will be refreshed when user manually triggers refresh
   • Automatic - Segment will be refreshed periodically. You can configure interval (e.g., Daily, Weekly, Monthly) and timing (e.g., 8AM).
4. Specify Expirations settings. If you specified Automatically as refresh method, you will need to set an expiration date for a segment after which the segment will stop auto refresh.
5. Define the segment by adding Filter group.
   • You can have multiple filter groups, you can provide a short description for each filter group. Click + Filter group to add filter group.
   • Specify time range for your filter group, you can choose relative date range or absolute date range.
   • Set up filter by selecting filter type, specifying filter scope, comparison operator, and then provide a value or range of values. For example, to filter users who had made two purchases, you would add a filter and select User has done as filter type, purchase as event and Total number(PV) as filter scope, = as comparison operator, 2 a value.
   • Within a filter group, filters can be evaluated with AND or OR logic, Click on And or +OR to add filter per your need.
6. Click Save.

View segment details

1. Select a segment, click on the segment name or click on Actions>View Details.
2. In the segment details page, you can view the segment settings, user number, and percentage of total user, segment history, and user samples.
3. You can download the user list of a segment as a csv file by click on the Export button per each run history.

Filters

Below table describes the available filters to create a segment.

Filter type	Description	Filter scope	Comparison operator
User has done	Select users had done certain event.	Total number: total number of the selected event occurred Number of times per day: The total number of times this event occurred per day Number of consecutive days: The number of consecutive days this event occurred Distribution of days: The number of days on which users triggered this event Sum/Max/Min/Average/Distinct count by numerical event attributes	= != > < >= <= >= between
User has not done	Select users had NOT done certain event.	Total number: total number of the selected event occurred Number of times per day: The total number of times this event occurred per day Number of consecutive days: The number of consecutive days this event occurred Distribution of days: The number of days on which users triggered this event Sum/Max/Min/Average/Distinct count by numerical event attributes	= != > < >= <= >= between
User has done in sequence	Specify an event behavior sequence. Users will only be selected if they trigger these events in the specified order.	Within a session: user must done the selected events within a session Across sessions: user has done the selected events in one or multiple sessions Directly followed: The next event needs to be immediately after the previous one Indirectly followed: There can be other events between the specified step sequence
User is/User is not	Select users by user attributes.	User attributes	is null is not null = != in not in contains not contains
User in segment/User not in segment	Select users based on whether user belong to a segment or not.	Segments