We had to manfully correct a couple records in our data mart - but don't
have the luxury of changing the source OLTP tables.

What is the best approach to prevent the ETL to pick this change us as a SCD
entry and move back the data correction to be in line with the source
tables?

The only way is to make your ETL more complicated by introducing some
sort of suppression table and swapping out the final table/view
feeding your dimension so AS never sees this record.

You're basically building a filter or lens through which your OLAP
system sees your OLTP data, and changing the picture that AS sees.

On Sep 28, 2:58 pm, "Joe" <hortoris...@gmail.dot.com> wrote:
> We had to manfully correct a couple records in our data mart - but don't
> have the luxury of changing the source OLTP tables.
>
> What is the best approach to prevent the ETL to pick this change us as a SCD
> entry and move back the data correction to be in line with the source
> tables?

if you do updates in the ETL process.
you can flag your manually updated records by adding a "manual update"
column
now when you do an update, verify that the manual update is set to 0 else
don't update the record.

when you do a manual cleansing change the manual update content from 0 to 1.

if you want to not insert specific records. then you need to store the keys
you want to exclude anywhere in a table and in the ETL you have to do a
lookup and then remove these rows from the dataflow.

it's 1 solution, but can works for you.


"Joe" <hortoristic@gmail.dot.com> wrote in message
news:7F9E040E-2690-4800-AC53-564229204D5F@microsoft.com...
> We had to manfully correct a couple records in our data mart - but don't
> have the luxury of changing the source OLTP tables.
>
> What is the best approach to prevent the ETL to pick this change us as a
> SCD entry and move back the data correction to be in line with the source
> tables?