Elementary does not natively support anomaly detection directly between two tables (e.g., comparing values or metrics across two tables) as a built-in feature. However, there are recommended workarounds: - You can create a view or a derived table that calculates the metric of interest between the two tables (for example, the ratio of row counts or a difference in values), and then apply an Elementary anomaly detection test (such as column_anomalies) on that derived metric. This approach is suggested for use cases like monitoring the ratio of row counts between two tables or similar comparisons. The Elementary team has also mentioned plans to support more flexible anomaly monitors on custom aggregations or metrics in the future, but this is not yet available out of the box [Anomaly detection between two tables workaround](https://slack.com/archives/C02CTC89LAX/p1677193288425909). - For direct data comparison (e.g., flagging data differences between two tables), Elementary supports Python tests for BigQuery and Snowflake, and one of the example tests is specifically for comparing two tables. This can be useful for validating that a test table matches a production table after changes [Python test for table comparison](https://slack.com/archives/C02CTC89LAX/p1677803458771979). In summary, while direct anomaly detection between two tables is not natively supported, you can achieve similar results using custom views or Python tests as described above.