Question 1

Do I need Attio record IDs in Databricks?

Accepted Answer

No, and that is what makes this practical. Identity can be an email on people, a domain on companies, or any suitably unique attribute you already hold, including a custom attribute carrying your own account identifier. Companies and people get dedicated domain and email handling for exactly this. Where a column holds several values, each is checked in turn. Matching on a non unique attribute updates every record it matches, so confirm uniqueness first.

Question 2

Where can modelled data land in Attio?

Accepted Answer

Any object your workspace exposes as writable, custom objects included, since objects are discovered live rather than read from a fixed list. Lists are writable too, so a segment can be pushed into a named list with its entry attributes set in the same sync. Notes, meetings, call recordings, workspace members and deal stage history are readable but not writable. Modes are create, update, and create or update, with no delete.

Question 3

How should I shape the model?

Accepted Answer

One row per record you intend to write, with scalar columns. Attio attributes hold individual values rather than nested structures, so STRUCT, ARRAY and MAP columns are worth resolving in the model SQL before the mapping stage, otherwise a nested value has nowhere natural to land. Where an attribute genuinely holds several values, such as a multiselect, an array column maps onto it directly rather than needing separate rows.

Question 4

What happens on a large first load?

Accepted Answer

It paces itself rather than failing. Attio permits a lower rate for writes than reads, and Polytomic throttles below that ceiling, working through records in batches with retries and backoff on rate limited requests, so a large backfill queues and completes over time. After that, model syncs are differential, so later runs carry only what changed and stay small even when the source table is large.

Question 5

How do I connect Databricks to Attio?

Accepted Answer

Connect Databricks with a SQL warehouse hostname and a token, or an OAuth service principal, and connect Attio with OAuth, granting access to the objects you want to sync. Objects are then discovered from your workspace, so custom objects and lists appear without extra configuration. Model the data with one row per record, choose the object or list, and map an identity such as email or domain.

Connect Databricks with Attio

Connect Databricks and Attio with Polytomic

Matched on email or domain

Custom objects and lists

Relationships by lookup

Trusted by businesses worldwide

Step-by-step integration guide

Connect Databricks in Polytomic

Connect Attio in Polytomic

Create a sync between Databricks and Attio

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 Databricks with Attio?