Observability for deep BigQuery cost and performance insights
Rabbit provides near real-time, actionable cost and performance visibility across your estate. Drill from account rollups to individual jobs, catch forecast divergence, and act before the bill lands.
<24hrs
anomalies caught within hours, not days
100hrs+
engineering hours reclaimed / month
40%
improvement in performance
Visibility across cost and performance
Dashboards show last month's bill. Rabbit gives you live observability across cost and performance, and points to the query, label, or project behind every spike. Cost insights, anomaly alerts, and performance monitoring each have their own surface.
Near real-time cost anomaly detection
Catch spend spikes and linear drift near real-time, not on the monthly bill.
Rabbit learns your usage patterns, builds a forecast model, and notifies you when spend diverges. Tune sensitivity and mark false positives as your team learns what matters.
Project and service · account and label
Forecast-based spike and drift detection
Email and Pub/Sub alerting, sensitivity levels 1–5
Cost issues caught within hours, not days
Deep cost insights
See where BigQuery spend actually goes, from account rollups down to individual jobs.
Rabbit joins billing export with job metadata so you can compare on-demand vs capacity cost, slot usage, and bytes processed per query.
Cost and slots by query, project, label, and reservation
On-demand vs capacity breakdown per job
Dbt/Airflow label enrichment and unlabeled resource highlights
50 engineering hours saved on average
Weekly and monthly digests
Keep finance and platform teams aligned without logging in every day.
Scheduled email summaries roll up cost trends, flagged anomalies, and top drivers for the cadence you choose.
Weekly or monthly delivery
Highlights anomalies and largest cost movers
Complements live alerts and in-product views
Performance Dashboard
Turn "BigQuery feels slow" into specific queries and contention failures.
See how queries perform by account, project, and label. Surface jobs that failed from slot contention and pinpoint the largest duration regressions.
Performance by query, account, project, and label
Slot contention failure visibility
Biggest duration regressions ranked and drillable
40% improvement in performance
Performance Alerting
Get notified when pipelines slow down or slot contention spikes.
Rabbit extends anomaly detection to performance signals. Same alert channels you use for cost, pointed at contention spikes and runtime regressions.
Slot contention spike detection
Pipeline run-time regression alerts (dbt, Airflow)
Anomalies caught near real-time
Detection without disruption
Rabbit surfaces drift with actionable insights and direct recommendations, so your team knows exactly what to fix and where to start. For certifications, residency, and how Rabbit handles access, see Security.
For the teams that run BigQuery
Same platform. Different priorities. Pick the path that matches your role.
Data Platform Leaders
Data Engineers
FinOps & Finance