Event pattern ETL process

To process large amounts of Unica Interact event pattern data and to make that data available for queries and reporting purposes, you can install a stand-alone Extract, Transform, Load (ETL) process on any supported server for optimal performance.

In Interact, all event pattern data for a given AudienceID is stored as a single collection in the runtime database tables. The AudienceID and pattern state information is stored as a Binary Large Object (BLOB). To perform any SQL queries or reporting based on event patterns, this new ETL process is necessary to break up the object into tables into a target database. To accomplish this, the stand-alone ETL process takes event pattern data from the Unica Interact runtime database tables, processes it on the schedule you specify, and stores it in the target database where it is available for SQL queries or additional reporting.

In addition to moving and transforming event pattern data to the target database, the stand-alone ETL process also synchronizes the data in the target database with the most current information in your Unica Interact runtime database. For example, if you delete an event pattern in the Unica Interact runtime, that event pattern's processed data is removed from the target database the next time the ETL process runs. Event pattern state information is kept up to date as well. So the information stored about event patterns in the target database is solely current data, not historical information.