Running aggregates have a similar functionality to customer aggregates, except that they do not define customers, but events. Running aggregates work with timeline, and consider all events preceding each specific event for which they are calculated.
Running aggregates can be created in Data & Assets, management section.
In running aggregates you specify desired information from the events, namely minimum, maximum, average, sum, count, if it exists for the customer, or information about first or last event described that match the filters. There is also an option to include or exclude the event itself from the calculation.
Running aggregates can only search in their respective past, and not in the future. Running aggregates are not calculated for a specific event type, and can be used for any event type.
Running aggregates are used to filter only events depending on their history in customer’s behavior. Running aggregates are used like other event attributes when being selected from a list.
Few of the uses are below:
- Getting information about last session start, e.g. UTM parameters, for events such as purchase
- Counting page visits, sessions or some other interesting events, so you can target customer on their nth occurrence of an event.
- You can execute scenarios when customers reach a certain threshold of e.g. revenue or some other score by defining a running aggregate using sum