BoostKPI is a ML-based platform that helps businesses find insights faster, better, and cheaper. Our key features include:
- Simple integration with cloud data warehouses (like Google BigQuery, Amazon RedShift, Snowflake and Databricks) or other SQL databases (like Postgres, MySQL and Clickhouse).
- Waterfall charts, heatmaps, pivots, and drilldown visualizations for root cause analysis. Support for saving and sharing analyses, as well as receiving their periodic snapshots.
- Personalized alerts for each stakeholder via email/slack/webHook/text.
- Users can provide optional feedback on the alerts to improve the signal-to-noise ratio.
The application has 4 key areas:
Use this tab to customize and access your “dashboard-views”, investigate the root cause of any change in KPIs compared to any baseline, and save any investigations as “snapshots”. “Snapshots” can be shared amongst team members.
Use this tab to view anomalies that resulted from any of your configured alerts. These anomalies, besides being routed to users or to slack/teams, can also be viewed on the dashboard.
Use this tab to set up granular alerts that combine unsupervised learning with business rules. The anomalies resulting from these alerts can be routed intelligently based on both the root cause as well as the magnitude of the deviation. The alerts can also be sent via slack.
Use this tab to connect to the cloud data warehouse.
An overview video of the root-cause analysis page is available at https://boostkpi.com.
To see configuration options for setting up alerts, please look at our blog post. It also shows an example of a rolling 7-day alert for a daily dataset.