What is the best place to implement Great Expectations

In the spirit of this question asked ~ 3 years ago - I am wondering if something has changed in the general reasoning since then.

Basically I am interested in some thoughts, advantages vs. disadvantages and especially practical experiences, nice moments vs. gotcha’s when implementing Great Expectations:

  • within the data pipeline (e.g. as events come in from the website, at each step of data ingestion, transformation, producing facts tables etc.)
  • at the very end on the final data warehouse / facts tables

Would highly appreciate also thoughts on experience with integrating Bigquery ← → Great Expectations.

Thanks in front.