We’re releasing a dramatic set of improvements to anomaly detection in Custom SQL Monitors. This will be released to the customer base in waves, starting tomorrow. It will be generally available to all customers in about 2 weeks.

Specifically, we’re taking many of the improvements we made to Metric monitors last year, and applying them to Custom SQL monitors that are configured to use automatic thresholds. Those monitors will see:

• Significantly tighter thresholds
• Support for “seasonal” patterns (day-over-day, week-over-week)
• Faster training of thresholds

We expect this to increase the number of alerts generated by Custom SQL monitors with automatic thresholds. If it becomes too noisy, we suggest users adjust the sensitivity of the monitor.

This week, we released two new options for how to apply Table Monitors from Settings. These additions provide more flexibility to select the desired tables. Specifically, we’ve added:

• Include all tables in a Domain: This is useful if you want to apply table monitors to all tables in a given domain.
• Exclude all tables in a warehouse, database, or schema: This is useful to include all tables in a warehouse or database, but exclude certain databases or schemas therein.

Dynamic schedules (When a table is updated) are now available for Metric, Validation, Comparison, and Custom SQL monitors. This scheduling option allows you to reduce time to detect data quality issues by running monitors shortly after the underlying table is updated.

We’re excited to share that we are evolving our legacy “insights” reports into a new, richer Data Exports product.

Originally launched as an experimental way to surface metadata insights, the legacy insights reports have grown misaligned with how our customers think about and interact with Monte Carlo. Over time, customers have shared that the current insights are difficult to use, hard to map to their mental model of Monte Carlo, and are missing key fields.

That’s why we’re transitioning to a new data exports that:

• Makes critical Monte Carlo objects—Monitors, Alerts, and Assets—exportable in clean, consistent formats
• Simplified and reliable exports that mirror what you see in Monte Carlo
• Accessible to download directly from the relevant pages in the UI

See more information in our data exports documentation.

Migration from Legacy Insights

We're sunsetting the legacy insights. Here's what to expect:

• The Insights tab will be removed from the UI
• The “Monitored Tables” insight has been moved to Settings → Table monitors
• Full deprecation will be July 1, 2025. If you have concerns about losing specific insight reports, please reach out to your customer success manager or [email protected]

We've updated the audience creation form to simplify the experience of configuring recipient channels and managing monitor notifications.

Previously, new users in Monte Carlo would land on the Alerts page. This change only affects go-forward new users. If your first login was earlier than today, you will continue to land on Alerts.

User research indicated that the majority of new users were seeking for information relating to their table, i.e. check lineage, see freshness or volume, add a monitor, etc. Landing on the alerts page often made it tricky to fulfill that intent, because it was hard to search & open for the specific table you cared about.

You can now acknowledge and resolve alerts, mark them as incidents, assign owners, and snooze monitors without leaving Teams or the Monte Carlo alert feed. With direct links to add Jira issues and view monitor details, you can get data operations work done quickly.

The feed is gradually rolling out to customers and Microsoft Teams will be released to all on Monday. If you would like it turned on now, please reach out to [email protected].

Investigating data issues just got easier with invalid row sample profiling. Quickly identify correlations with at-a-glance statistics and top values. Drill in deeper, copy entire columns directly to your clipboard, customize views by column selection and sorting, or download rows as a CSV for offline analysis.

Now available in public preview—rolling out to accounts over the next week. If you have feedback or requests on this, we'd love to hear more.

We've added support for bidirectional synchronization for alert priority, owner, status, and comments between Monte Carlo and Opsgenie so that users no longer have to reconcile the information manually.

A user with this role has all the same privileges as an Editor, except they cannot:

1. change table monitoring rules
2. edit asset metadata, such as Asset Tags or Description.

This means that a Monitor Editor can create monitors like Metrics and Validations, but cannot add Table Monitors that may contribute to your Monte Carlo bill. See docs for more details.