In my testing, the answer to this is no, which makes sense. Wondering if there are alternatives? Performance is significantly improved by batching, but the tracked field data is really nice to have.
As you observed, with a batched response from Salesforce (or any connector), the tracked fields operate at the document level so you wouldn't be able to track at the logical record level. I don't have any clever yet practical workarounds for this situation--sorry.
Out of curiosity what kind of performance difference are you seeing with/without batching and for how many records?
How to capture field level changes in SF
Hey Adam Arrowsmith
Thanks for the reply.
I don't have the exact numbers anymore, but it was in the neighborhood of 10k records with a decrease in runtime of ~30-40% if memory serves, maybe more.
Retrieving data ...