There have been instances where after a few days we identified that something has gone wrong in the ELT of a table in a scheduled run. Now we don’t know what caused the duplicates or the missing entries, or in some cases it was a DML query gone wrong, and the only solution that we could think of was to do a full refresh of the table gone wrong. We don’t want to do a full table update for all the tables since it would consume a lot of time and money. What would be the best way to do that (on target-bigquery if that matters)?

Can you share any relevant part of your config and explain how you are using it?