End-user augmenting data before ingesting to DW a.k.a ad-hoc ETL — is that a thing?

Hello Reddit, 

with my university classmates, we are looking into the limits of current ETL technologies. After our initial research, we found out that there aren’t many solutions that would allow end users to create ad-hoc ETL processes. In other words, the ETL process cannot be fully automated and the end-user has to do data manipulations (such as mapping columns, groping data by some attributes etc..)

On the one hand, we feel that similar use cases must be a common …

ad data datascience etl