Hello @VaderKB 

You're right that OPTIMIZE makes reads faster by reducing the number of files. For writes using append mode, it doesn't directly speed up the operation itself. However, having fewer, larger files from a previous OPTIMIZE run can improve the overall performance of subsequent reads that might be part of a larger job, which could make the entire pipeline seem faster.

You can go in detail from this document 
https://www.databricks.com/discover/pages/optimize-data-workloads-guide

as the issue is resolved can you please close the case by selecting a solution from this page?

Thank you.