Question 1

Does change data capture work end to end into MotherDuck?

Accepted Answer

Yes. SQL Server maintains its own change tables once CDC is enabled per table, Polytomic reads from those, and the records load through MotherDuck's Parquet staging path. Staging is transport only: rows land in a staging table and are then reconciled with update and insert statements, so an incremental run genuinely applies changes rather than appending another copy of everything.

Question 2

Do deleted rows disappear from MotherDuck?

Accepted Answer

They are marked rather than removed, which is usually what you want for analysis. When the table has a primary key, deletes captured by SQL Server CDC are recorded in MotherDuck with a deletion timestamp column so history is preserved and you can filter them out in queries. Hard delete propagation is available if you would rather the rows were physically removed. Without a primary key the sync falls back to append and deletes cannot be tracked.

Question 3

What happens when the SQL Server schema changes?

Accepted Answer

This is the operational cost to plan for. SQL Server CDC captures the columns that existed when capture was enabled on a table, so a new or altered column does not appear in the change stream on its own. Picking it up means rebuilding the capture instance for that table and resyncing it into MotherDuck. If your source schema is stable this rarely comes up; if it changes often, factor it into which tables you capture.

Question 4

Can Polytomic reach a SQL Server that is not publicly accessible?

Accepted Answer

Yes, through an SSH tunnel using a bastion host, port, user, and private key. That matters more here than with cloud native sources, since so many SQL Server instances sit on premises or inside a private VPC. It is frequently what makes this pairing possible at all: the database stays where it is, and the analytical copy lives in MotherDuck where people can query it without network access to production.

Connect MotherDuck with Microsoft SQL Server

Connect MotherDuck and Microsoft SQL Server with Polytomic

Native change data capture

Deletes land correctly

Reaches private instances

Trusted by 1,000+ businesses worldwide

Step-by-step integration guide

Connect MotherDuck in Polytomic

Connect Microsoft SQL Server in Polytomic

Create a sync between MotherDuck and Microsoft SQL Server

Map fields and enable the sync

Frequently asked questions

The sync and ETL platform that does it all

Powerful data sync

Advanced analytics

Enterprise security

Easy setup

Ready to connect MotherDuck with Microsoft SQL Server?