Question 1

What does Customer.io accept from a Redshift model?

Accepted Answer

People and their attributes, custom objects, segments, person events and people devices. That covers what a messaging tool needs to know: who somebody is, what they did, and which group they belong to. Devices are append only, so they are added rather than matched and updated. The rest follow normal matching, where you map a column from your model to the identifier Customer.io uses.

Question 2

Do attributes need creating in Customer.io first?

Accepted Answer

No. New attributes and properties can be declared straight in the mapping for people, events and custom objects, so pushing a new column from your model needs no setup on the Customer.io side. Adding something like a churn score or plan tier becomes a change in one place rather than a coordination exercise across two tools.

Question 3

Does Redshift Serverless work as the source for this?

Accepted Answer

Yes, though the connection differs. Provisioned clusters use direct database access with a username and password. Serverless uses either the Redshift Data API or a direct Serverless endpoint with temporary IAM credentials, and it enables join pushdown. Reads are ordinary SQL either way. Two things to know: the Data API can throttle under heavy or highly concurrent load, where the endpoint method tends to suit better, and IAM permissions do not replace the database USAGE and SELECT grants, which still have to be mapped to a database role.

Question 4

What if a person in the model is not in Customer.io yet?

Accepted Answer

The sync mode decides. Update or create makes them, update only skips them, and create only makes the new ones while leaving matches alone. Choose based on whether Customer.io or the warehouse should introduce new people. If a related record referenced in a mapping cannot be found, Customer.io can optionally ignore that relationship and still write the base record instead of failing it.

Question 5

How do I avoid maintaining the same segment logic twice?

Accepted Answer

By defining it once, in the warehouse, and letting reverse ETL carry the result. Anything depending on usage trends, account health or churn risk is built from product, billing and CRM data that a messaging tool cannot see, so rebuilding it inside Customer.io means two versions that slowly disagree. That drift is how a campaign ends up messaging people your reporting already counted as gone.

Sync Redshift to Customer.io

Sync Redshift to Customer.io with Polytomic

People, events and objects

Declare attributes in the map

No bucket on the read side

Trusted by businesses worldwide

Step-by-step integration guide

Connect Redshift in Polytomic

Connect Customer.io in Polytomic

Create a sync between Redshift and Customer.io

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 Redshift with Customer.io?