May 2026: Auth0 SSO, Salesforce Formula Sync & Ad Reporting
ELT & CDC
May's release cycle focused on deepening ad platform reporting coverage, expanding Salesforce sync control, and hardening pipeline reliability across connectors and notification workflows.
Introducing Auth0 SSO
Users can now sign in via SSO using Auth0, and organization admins can configure SSO and enforce IdP-only logins for their accounts, also giving teams a centralized, secure authentication path without managing individual credentials.
Salesforce Formula Field Sync
Salesforce formula fields can now be synced either globally for the entire source or on a per-table basis, giving teams granular control over which calculated fields are included in their pipelines. Formula fields are clearly marked in the UI, so users can identify and configure them without additional documentation lookups.
New Features
- Snowflake Custom Host: Users can now specify an optional custom Snowflake host when configuring a Snowflake destination: supporting non-standard deployments and use cases like those of Yuki Data, a partner who's focused on optimizing data warehouse cost and spend on a custom Snowflake setup.
- Google Ads Time-Variant Keyword and Budget Reports: Five new KEYWORDS_PERFORMANCE_REPORT streams (daily, weekly, monthly, quarterly, and yearly) are now available, along with a new BUDGET_PERFORMANCE_REPORT_BY_CAMPAIGN collection, making time-segmented keyword and budget data accessible for more granular paid-search analysis.
- Google Ads All-Time Metrics Collection: A new `_TIMES` collection is now available for every Google Ads report, consolidating all-time aggregated metrics alongside the existing time-windowed streams - removing the need to union multiple reports to get full-range totals.
- TikTok Ads Expanded Field Coverage: A broad set of additional fields is now available on ad_groups, ads, campaigns, creative_assets_videos, and pixels collections, with metrics and dimensions exposed on most report collections and dimension and metric selections persisted across sessions.
- Destination Column Default Preservation: Existing column defaults on Redshift and Snowflake destination tables are no longer overwritten during pipeline runs, protecting intentional database-level defaults from being cleared by sync operations.
Improvements
- When new tables are added to a Salesforce V2 source, their incremental replication key and streaming support are now saved automatically, so incremental syncs and streaming options work correctly for newly added tables without additional configuration.
Fixes
- Chunking issue on composite string primary keys that could cause rows to be missed or duplicated during chunk-based syncs - now resolved.
- Snowflake configuration validation no longer rejects pipelines where the custom host field is unset, allowing existing Snowflake pipelines to start normally.
- PostgreSQL 14+ tables that had not been analyzed no longer stall on resync - reliable initial sync behavior has been restored.
- Pipeline hooks no longer send duplicate email notifications when the same hook is attached to multiple pipelines.
- Enterprise SAML users are no longer blocked at login when their identity provider does not set the `email_verified` attribute.
- Data Observability sync lag alert no longer triggers a false positive when a sync returns no data.
Performance Enhancements
- DB2 and BigQuery connectors now support parallel chunk workers for initial sync, splitting large tables across concurrent threads and reducing time-to-first-value when onboarding new sources or backfilling historical data.
Let us know what you think of these updates - your input shapes what we build next.
3 days ago