Aggregates
An aggregate is a summarized data about a selected event from a specified time range. You can determine the desired output – whether it is the latest or first value for a particular event parameter, median, average, sum, minimum, maximum, and more.
In Synerise, you can create the following aggregate types:
- Profile - Profile aggregates include customer-context, allowing for displaying the result for individual customer. Within the aggregate preview, you can select a customer identifier to view the results tailored for that specific customer.
- Event - Event aggregates let you analyze an occurrence of an event and occurrences before it. These aggregates act like custom event parameters which can be used in filters (as a property or value). Event aggregates can be applied to filters with different events than the event in the aggregate, which can help you create complex analyses.
Aggregates serve more than just analytical purposes; they are essential components within larger use cases and complement other Synerise features, such as Predictions and AI recommendations. The table below showcases what you can do with aggregates:
| Purpose | Practical examples | Aggregate type |
|---|---|---|
| Simple analysis | - Discover a customer’s preferred channel - Discover a customer’s preferred category - Calculate duration of the latest session |
Profile |
| Complex analysis | - RFM analysis - Dashboard for a call center that contains transaction statistics (for example, name of the product purchased the most recently, transaction value, number of transactions, and so on) |
Profile |
| Providing context for AI features | - Recommendation of similar products filtered by favorite category of a customer (the favorite category is provided by the aggregate result) - Recommendation boosted by the results of AI prediction for best brand for a customer (results of the prediction are retrieved from an aggregate) |
Profile |
| Creating custom event parameters | - Gather a group of customers whose average transaction value exceeds $500 after second visit in the store - Calculate recommendation revenue after the first click within 24 hours |
Event |
