Event time processing - time base aggregations

Timestamps are important because they impact the results of your aggregation.

What’s powerful with KSQL is that supports time-based aggregations which allows you to ask questions such as:

  • How many events occurred in the last five minutes?
  • What was the total value of orders placed each hour today?
  • How many minutes have elapsed between customer actions?

So having a good handle on the timestamps in your streams is critical to get the correct results. Read more from Confluent here on how to identify & use the right timestamp for your use-case:
https://www.confluent.io/stream-processing-cookbook/ksql-recipes/event-time-processing