Alias Tables Are Now Materialized as Database Views

We've completed the rollout of materialized aliases to all Keboola customers. Alias tables are now backed by real database VIEWs in both Snowflake and BigQuery, bringing them fully into your warehouse as first-class objects rather than metadata-only references.

The rollout landed across all tiers between May 22 and May 28, 2026: PAYG GCP and Azure first, followed by multitenant projects. Single-tenant deployments will follow between June 1 and June 4, 2026. There are several projects that do not have this feature enabled yet, and we will update them individually.

What changed

• Aliases are now database VIEWs: Every alias table is materialized as a VIEW in Snowflake or BigQuery, making it a real object in your warehouse schema — not just a Keboola-internal reference.
• Visible in the SQL editor: Aliases appear in the SQL editor through read-only storage, so you can query them alongside your regular Storage tables without any extra configuration.
• Shareable across projects: Alias tables can now be shared across projects, enabling cross-project reuse of curated datasets without duplicating data.

Why it matters

Previously, alias tables were a useful organizational construct, but they were invisible to tools that connected directly to the warehouse backend. By materializing them as VIEWs, they now show up in any SQL client, BI tool, or data app that queries your warehouse — no Keboola-specific abstraction required. This makes aliases a practical, shareable building block for curated data products.

If you use alias tables in your projects, you should start seeing them in the SQL editor automatically. No configuration changes are required on your end. If you run into anything unexpected, please don't hesitate to submit a support ticket — we're here to help.