Event segmentations work similarly to customer segmentations. They are used to create custom groups of events based on various event attributes. Event segments are created for one specific event, and cannot combine groups of different events. Their purpose is to better categorize events for further use in other analyses
Creating an event segmentation
Event segmentations are created in the Data & Assets, management section. First step is to select the event which is to be categorized. That will stay identical for all segments. For specifying the events, any attribute or previously created running aggregate or expression can be used.
Each event can be only in one category, but does not need to be in the segmentation at all. Segments are filled from left to right, so if event already belongs to the first segment, it cannot be in any of the others. If two identical segments are created, one will contain all the events that match the condition, and the second will be empty.
There is currently no OR filter for the events in segment, except for the operators like “in”, etc. Adding multiple attributes serves as AND for the filters. You can however use event segmentations as well when creating a new segmentation, resulting in ability to use multiple segments with different rules to achieve the desired effect.
Event segment will NOT be displayed as an attribute for events in customers. It can be worked with as with any other attribute and will be found under event segmentations section of the attributes.
Event segmentations can be used when there a need to group some attributes with many different values to fewer categories. This is either with numeric values, or when multiple different values share the logic. The event segmentations are usually used in other analyses for better understanding of the data. It usually helps with charts that are then becoming too hard to read. Examples below:
- For example facebook, fb, googleplus and/or any others can be grouped into one category “social media”, google, bing and other search engines can be grouped into “search engines” etc.
- Grouping of purchases into small, medium and big based on total order price