ML anomaly detection.
Tools that learn statistical baselines for tables and surface deviations without pre-defined assertions.
What counts as ML anomaly detection?
Manually-authored tests can only catch failure modes someone thought to write. ML anomaly detection learns what "normal" looks like for each table — row counts, freshness intervals, value distributions — and alerts on deviations. Trade-off: requires a training window (typically 14–30 days) before the signal stabilizes, and highly seasonal data tends to produce false positives until the model has seen enough cycles.
10tools, grouped by primary cluster.
Acceldata
Acceldata
Enterprise data observability with ML data quality, reconciliation, and a built-in catalog — strong on hybrid and on-prem estates.
Anomalo
Anomalo
GUI-first ML anomaly detection at petabyte scale — pivoting in 2026 around agentic AI and unstructured-data monitoring.
Bigeye
Bigeye
Enterprise data observability with Autometrics ML thresholds — repositioning in 2026 as an AI Trust Platform with runtime governance.
Elementary
Elementary Data
The dbt-native observability layer — tests, anomaly detection, and lineage that live inside your dbt project.
Metaplane
Metaplane (Datadog)
ML-powered, no-code data observability for the dbt and warehouse stack with automatic column-level lineage — now Metaplane by Datadog.
Monte Carlo
Monte Carlo Data
Warehouse-side data observability for teams whose problems are upstream of dbt — ingestion, streaming, and across the full pipeline.
Sifflet
Sifflet
EU-built full-stack data observability pairing ML-driven monitoring with an embedded catalog and field-level lineage.
Soda
Soda Data
YAML-first data contracts and observability — SodaCL plus Soda Cloud, with anomaly detection and a self-hosted Kubernetes runner.
Collibra
Collibra
Enterprise data-and-AI governance incumbent: catalog, glossary, workflow stewardship, lineage, and a separate ML data-quality module.
DataHub
Acryl Data
Apache-2.0 metadata platform with a serious managed counterpart — strongest event-driven architecture and column-level SQL lineage in OSS.
Head-to-head, side by side.
Drill into a different capability.
How this list is built.
Inclusion here is one boolean on each tool's structured profile — if a tool you'd expect is missing, the field is recorded false or not yet verified, never an editorial call. See the methodology for how each field is sourced.