Fivetran Release Notes

We fixed an issue where rows were incorrectly dropped from the EMAIL_RAW_EVENTS, UNSUBSCRIBED_EMAILS tables when a non-essential key field was blank, despite each row having a valid unique identifier. Affected tables will automatically resync to restore any missing rows.

The Typeform connector can now sync data from accounts hosted in the EU region. To support this improvement, we have added a new drop-down menu, Region, to the connection setup form, where you can select the region of your Typeform account.

We have added the following new tables:

• ENGAGE_FLOW_PROSPECT_ASSIGNMENT
• ENGAGE_FLOW_STEP
• TASK
• TASK_ACTION
• TASK_CONTACT_EMAIL
• TASK_CONTACT_OPT_OUT_CHANNEL
• TASK_CONTACT_PHONE

We have improved sync reliability for the PROJECT_VULNERABILITY table. We now re-import this table once a week to capture updates and deletes. We are gradually rolling out this change to all existing connections.

We have improved sync reliability for the PORTFOLIO table and its child tables. In addition to incremental syncs, we now re-import these tables once a week to capture updates and deletes.

Reddit Ads no longer supports certain breakdown combinations. A reporting request may include no more than one of the following breakdowns:

• KEYWORD
• COMMUNITY
• INTEREST

In addition, KEYWORD, COMMUNITY, and INTEREST breakdowns cannot be combined with any of the following breakdowns in the same request:

• DMA
• REGION
• GALLERY_ITEM_ID
• OS_TYPE
• COUNTRY

We now support syncing Journals data for Xero connections. To support this enhancement, we have added a new toggle, Sync Journal Data, to the setup form. New Xero connections created after September 22, 2026, using Web app can enable this option during configuration. This feature is available only for Xero accounts on the Advanced or Enterprise tiers. We are gradually rolling out this feature to all existing connections. For more information, see our setup instructions.

We've added a new records_extracted log event. The event reports the number of records extracted from the source for a table, along with the destination schema and table names.A single sync can generate several records_extracted events for the same table. Each event counts only the records extracted since the previous event for that table, so sum the count values to get the total for a table in a sync.

The sync_end log event now includes the load_volume_b field, which reports the volume of data loaded into your destination during the sync, in bytes. The field is only present for successful syncs of connections that report load volume.

You can now use Quickstart data models with Managed Data Lake Service destinations. This functionality is in Beta.

We have added the following tables:

• AUDIT_LOGS to capture administrative and user actions performed in your Kustomer account
• AUDIT_LOG_ATTRIBUTE_CHANGE to capture attribute-level changes associated with each audit log entry
• AUDIT_LOG_RELATIONSHIP_CHANGE to capture relationship-level changes associated with each audit log entry

We have added two new tables, CUSTOM_OBJECT_FIELD and CUSTOM_OBJECT_RECORD. The CUSTOM_OBJECT_FIELD table contains the field definitions for your custom objects, so you can understand the structure of your custom data. The CUSTOM_OBJECT_RECORD table contains the individual records created for each custom object type you define in Zendesk, so you can analyze custom business data alongside your support tickets.We have added a new table, POST_CONTENT_TAG, to store the content tags associated with each community post. We are gradually rolling out this change to all existing connections.

We've added a new records_extracted log event. The event reports the number of records extracted from the source for a table, along with the destination schema and table names.A single sync can generate several records_extracted events for the same table. Each event counts only the records extracted since the previous event for that table, so sum the count values to get the total for a table in a sync.

The sync_end log event now includes the load_volume_b field, which reports the volume of data loaded into your destination during the sync, in bytes. The field is only present for successful syncs of connections that report load volume.

We now sync future data up to 12 months from the current date for the EMPLOYEE_TIME_AWAY_FROM_WORK table.

We have changed our sync strategy for the MANAGER_REVIEW table. In addition to incremental syncs, we now re-import this table once a day to capture updates and deletes.

All new connections set up on or after October 5, 2026 preserve the original JSON representation when storing JSON values in STRING columns. Previously, JSON string values such as "true", "null", and "123" could lose their enclosing quotes, potentially changing their meaning in the destination.

We have made the following schema changes as part of the migration of Greenhouse connections from Harvest API V1 and V2 to V3:

• Added the _V_1 suffix to all table names that contain data from Harvest API V1 and V2
• Removed the _V_3 suffix from all table names that contain data from Harvest API V3

We will continue syncing tables with the _V_1 suffix for as long as Greenhouse provides data through Harvest API V1 and V2. Be sure to update your downstream queries to align with these changes. Failing to do so may lead to missing or incomplete data in your data pipelines.