Question 1

How does writing into MongoDB work?

Accepted Answer

As a model sync with mapped document fields and paths, not as bulk table replication, because MongoDB is not available as a bulk sync destination. In practice that means you decide the document shape you want and map Snowflake model columns onto it, including nested paths. The collection can be created for you, and arbitrary document fields can be added, so the destination structure does not have to exist in full beforehand.

Question 2

How are existing documents matched?

Accepted Answer

By the document identifier when you map one, which is the most reliable option, and serialized ObjectID values are converted rather than being treated as plain strings. If you do not supply an identifier, matching falls back to filtering on whichever destination field you nominated as the identity. Classification into insert or update work happens before the write, so updates are issued without relying on an upsert flag.

Question 3

Is there anything to watch on data types?

Accepted Answer

Datetimes, mainly. Values mapped to a datetime field are strictly coerced, so anything malformed becomes a record level error rather than being written as a string or silently dropped. That is the behaviour you want, since it surfaces the specific rows, but it does mean a model emitting loosely formatted timestamps will produce errors on the first run. Normalising those in the Snowflake model is quicker than debugging them afterwards.

Question 4

Why push warehouse results into a document store?

Accepted Answer

Because the application reads from Mongo and needs values it cannot compute itself. Segment membership, propensity scores, aggregated usage and recommendation sets are all natural to model over warehouse data and awkward to derive inside an application database. Writing them back as documents means the app reads them the same way it reads everything else, with no extra query path and no dependency on the data warehouse being available at request time.

Connect Snowflake with MongoDB

Connect Snowflake and MongoDB with Polytomic

Mapped document fields

Matches on document ID

Grouped bulk writes

Trusted by businesses worldwide

Step-by-step integration guide

Connect Snowflake in Polytomic

Connect MongoDB in Polytomic

Create a sync between Snowflake and MongoDB

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 Snowflake with MongoDB?