Question 1

What Salesforce data can I load into Databricks?

Accepted Answer

Whatever the connected user can see, which covers accounts, contacts, leads, opportunities, cases and campaigns along with your custom objects. Formula and rollup summary fields arrive as calculated values rather than as formulas, so the logic does not have to be rebuilt in SQL or in a notebook. Field level security still applies, so a field hidden from the connected user will not appear.

Question 2

Does Unity Catalog change how tables are managed?

Accepted Answer

Not the loading path, which is stage then COPY INTO then merge either way, but it does change how a table gets replaced. Unity managed tables can be rebuilt through a temporary table followed by a drop and rename. On the default hive_metastore path the final table is replaced in place instead, because rename carries caveats there. One thing that does depend on Unity: choosing an external destination location is bulk sync only and requires Unity Catalog.

Question 3

How does Polytomic keep Salesforce data current in Databricks?

Accepted Answer

After the first full load, each run asks Salesforce only for records changed since the last successful run. Those stage to your storage, load into a temporary table with COPY INTO, then MERGE INTO the destination on your configured key so existing rows update in place. Salesforce does not broadcast schema changes, so new fields are found by polling rather than being pushed, and newly discovered columns are added with an ALTER TABLE.

Question 4

Are deleted Salesforce records reflected in Databricks?

Accepted Answer

Yes, within Salesforce's limits. Reads include archived and soft deleted records and expose the deletion flag, and the merge records a deleted at timestamp so you can count what has gone rather than losing it silently. Enabling propagate hard deletes removes those rows after the load. Records purged from the Recycle Bin are gone at source and need a full resync to reconcile. Note that this hard delete option is a bulk sync feature, not something model syncs expose.

Question 5

Why load Salesforce into Databricks specifically?

Accepted Answer

Because CRM data is usually one input to something larger. Attaching an opportunity to product usage or support history needs a join, and if the same lakehouse already holds event data and feeds model training, the CRM belongs there rather than in a separate reporting copy. ETL also gives you a queryable history rather than fields that overwrite themselves, which matters for anything retrospective.

Sync Salesforce to Databricks

Sync Salesforce to Databricks with Polytomic

Custom objects included

Unity Catalog aware

Keyed MERGE INTO

Or sync Databricks to Salesforce

Trusted by businesses worldwide

Step-by-step integration guide

Connect Salesforce in Polytomic

Connect Databricks in Polytomic

Create a sync between Salesforce and Databricks

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 Salesforce with Databricks?