Question 1

What Gong data can I load into Databricks?

Accepted Answer

Calls with their metadata, participants, context, topics, trackers and tracker occurrences, plus interaction data. Separately there are transcripts, keyword tracker definitions, scorecards and answers, users, workspaces, library folders and contents, and Engage flows. Tracker hits are often the most immediately useful, because a hit is a countable structured signal where a transcript is text you still have to process.

Question 2

Are the tables Delta, and can I set table properties?

Accepted Answer

New tables are effectively Delta, relying on the workspace default table format rather than an explicit USING DELTA in the generated DDL. Delta properties can be applied, including Change Data Feed and retention settings, and UniForm properties are available where the catalog is Unity backed. Existing tables are not converted, so if you point a sync at something already there in another format, it stays as it is.

Question 3

How much of Gong can be read incrementally?

Accepted Answer

Only calls and transcripts, and only if you turn it on. Those use a time watermark with a rolling seven day lookback, which exists because a transcript is not ready the moment a call ends. Everything else is a full read. Be clear about the limit: this is not update complete change capture, so a change landing outside the window can be missed. Where completeness matters, pair the incremental schedule with a periodic full read.

Question 4

Does Polytomic remove deleted Gong calls?

Accepted Answer

No, because Gong has no delete feed for this connector. Treat it as a snapshot source. Nothing produces a tombstone, so a call removed in Gong stays in your table until a full read reconciles it. That is why the incremental calls and transcripts path should not be described as delete aware even though it genuinely is incremental. If your reporting counts calls, schedule an occasional full read.

Question 5

Why put conversation data in a lakehouse?

Accepted Answer

Because a transcript is raw material rather than an answer, and processing it is exactly what this platform is for. Customers describe wanting Gong landed first so they can shape it, join it to accounts and opportunities, and only then act. ETL puts the tables within reach of both SQL and a notebook, so counting tracker hits and running language analysis over transcripts happen in one place.

Sync Gong to Databricks

Sync Gong to Databricks with Polytomic

Delta tables with properties

Calls, transcripts, trackers

Catches late transcripts

Or sync Databricks to Gong

Trusted by businesses worldwide

Step-by-step integration guide

Connect Gong in Polytomic

Connect Databricks in Polytomic

Create a sync between Gong 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 Gong with Databricks?