Question 1

How does the HubSpot incremental read work?

Accepted Answer

The first ETL run is a full scan through HubSpot's list API. After that Polytomic searches on lastmodifieddate, collects the IDs that changed, then batch reads the properties you selected. HubSpot's search index lags slightly behind its writes, so consecutive runs overlap their lookback windows and deduplicate. That is why a record can be fetched twice and still written once.

Question 2

Do deletions in HubSpot reach Postgres?

Accepted Answer

Not through the incremental path. The search API only returns records that still exist, so a deleted record stops appearing and its row stays behind. Schedule a periodic full resync, which marks the missing rows through the deleted at column, or switch on hard delete propagation to remove them physically. There is also a newer option to include archived HubSpot records, but it is feature gated, so check whether it is turned on for your account.

Question 3

Can I get deal stage history?

Accepted Answer

Yes, as separate tables. Contact, company and deal property history are each their own bulk object, and every change becomes a row carrying the object ID, property name, new value, timestamp, source and the user behind it. That shape is what stage duration analysis needs, and it is one of the things ETL gives you that the HubSpot interface will not. Property history for custom objects is not exposed.

Question 4

Are custom objects and properties picked up?

Accepted Answer

Yes. HubSpot schemas are discovered automatically, custom objects and custom properties included, and the cached copy refreshes every twelve hours. Two things to know. A sync you already configured will not silently start writing a newly discovered field, you have to enable it. And custom objects read in full each run rather than through the search based incremental path.

Question 5

Should this land in my production database?

Accepted Answer

Worth deciding before you start. Teams often point this at the same instance their application runs on, then find CRM tables competing with production queries. A separate database, or an instance sized for analytics, keeps the two apart. Contacts and property history are the two tables most likely to surprise you on volume, so if you are staying on one instance, start with those disabled.

Sync HubSpot to PostgreSQL

Sync HubSpot to PostgreSQL with Polytomic

Property history as tables

Searches, does not rescan

Custom objects discovered

Or sync PostgreSQL to HubSpot

Trusted by businesses worldwide

Step-by-step integration guide

Connect HubSpot in Polytomic

Connect PostgreSQL in Polytomic

Create a sync between HubSpot and PostgreSQL

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 PostgreSQL?