Question 1

What HubSpot data can I load into Databricks?

Accepted Answer

The CRM objects covering companies, contacts, deals, line items, products, tickets, meetings, calls, emails, notes, tasks, invoices, owners and your custom objects. Around those sit property histories, associations, engagements, forms and submissions, lists and memberships, pipelines, marketing events, email events and campaigns, workflows and subscription changes. Custom properties are discovered as they appear and association labels you defined come through as fields.

Question 2

What happens when a HubSpot property changes type?

Accepted Answer

The old column is renamed to an archive name that carries its previous type, something like field_STRING, and a fresh column with the original name and new type is added next to it. Nothing breaks and no history is lost, which matters on a CRM where somebody occasionally changes a field from text to a number. New properties are simply added with an ALTER TABLE. Dropped source properties are kept during ordinary runs, so old columns linger until a rebuild removes them.

Question 3

How does Polytomic keep HubSpot current in Databricks?

Accepted Answer

For the main CRM objects the HubSpot search API finds records changed since the last run by last modified date, then a batch read collects full properties and associations for those records. Those stage to your cloud storage, load with COPY INTO into a temporary table, and MERGE INTO the destination on your key. The first load uses the list API instead, since there is nothing to compare against yet.

Question 4

Are deleted HubSpot records removed from Databricks?

Accepted Answer

Not by incremental runs, and it is worth designing around early. The HubSpot incremental API does not report deletions, so a record removed in HubSpot will not be flagged by a normal run. The options are full refresh mode or periodic full resyncs so reconciliation clears rows that are no longer present. A few entity types do expose an archived or deleted timestamp, but that is not general delete tracking across the CRM.

Question 5

Why load HubSpot into a lakehouse?

Accepted Answer

For history and for joins, and because the CRM is rarely the only input. HubSpot shows current state well but records nothing about what a deal was worth three months ago in a queryable form. ETL into a data warehouse gives you that history at your own grain and puts CRM records next to product events and billing data, which is where scoring and propensity work usually needs them anyway.

Sync HubSpot to Databricks

Sync HubSpot to Databricks with Polytomic

Property history included

Type changes archived

Custom properties found

Or sync Databricks to HubSpot

Trusted by businesses worldwide

Step-by-step integration guide

Connect HubSpot in Polytomic

Connect Databricks in Polytomic

Create a sync between HubSpot 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 HubSpot with Databricks?