In addition to the existing dbt test failures and model errors, we are now sending dbt warnings as MC incidents. Unless you have relatively noisy warnings, you should already start receiving dbt warnings the same way and same place you are getting failures for the same tests, without needing to configure anything.

To change the configuration for your warnings in MC, you can control go to Integrations -> Settings -> dbt to toggle the options. You can also opt in to group repetitive warnings into the same incident to prevent multiple alerts.

Our integration with Jira makes it easy to operationalize an incident management process. Users can choose to “push” an incident to Jira through a human-in-the-loop interaction, or they can directly send incidents to Jira through notifications. The status of those issues in Jira can then be automatically synced back to Monte Carlo, to close the loop.

Previously, only Jira Cloud was supported. But now, enterprises using Jira Data Center can integrate as well. Functionality between the Jira Cloud and Jira Data Center integrations is the same.

To learn more about our integration with Jira, and steps to integrate Jira Data Center, see our documentation.

Under the Usage UI, for all Orgs that have migrated to the new monitoring rules, a button to "Download monitored_tables.csv" will be available to download a timestamped csv. The download will include all current monitored tables at that point in time. Changes to the monitoring rules in the Usage UI will be immediately reflected in any subsequent downloads of the csv.

Columns included in the export:

• Integration
• Database
• Schema
• Table Name
• Type
• Importance Score
• Last Activity

Monte Carlo now delivers a personalized Weekly Data Reliability Summary email. This digest provides a convenient overview of critical data health metrics from the previous week.

Key Summary Sections:

• View incidents: A quick link into the Incident Feed for incidents raised in the past week.
• Incidents assigned to you: The 5 most recent Incidents by creation date, raised in the past week that are currently assigned to the digest recipient.
• Incidents & Response: Summary of key metrics around Incidents raised by Monte Carlo and how they were responded to.
• Top 5 assets with incidents: The 5 assets with the most incidents raised against it over the past week, sorted by total incidents raised for that period
• Top 5 monitors with most incidents: The 5 monitors with the most incidents raised over the past week, sorted by total incidents raised for that period.
• 5 most recent new monitors added: The 5 newly created monitors over the past week, sorted by creation date.

Availability: Account Owners, Domains Managers, Editors, Responders, and Viewers will start receiving this Weekly Digest automatically.

Manage Preferences: Opt in or out of the weekly digest within your User Profile settings.

For more details, refer to our documentation.

Formerly known as Field Metrics, the new experience for Metric Monitors provides more flexibility and functionality to do deep quality checks on a given table, including:

• Feature parity between metrics monitored using automated vs manual thresholds. Segmentation, scheduling options, and filtering options are now the same regardless of if a monitor is using manual or automated thresholds.
• Manual and automated thresholds can now be set using a single monitor. For example, I could monitor 3 field for Unique (%) with an automated threshold, and 5 different fields for Null (%) using a manual threshold, all within a single monitor.
• New UI to create Metric Monitors, emphasizing ease of use.

To learn more, check out our documentation on Metric Monitors.

In the Usage UI, there are now options to set Monitoring Rules at the Warehouse and Database levels. When a rule is set at one of these levels, all lower levels will show a read-only rule with references back to the higher level where it was initially configured. Additional rules can be added in include/exclude at lower levels and are OR'ed with the inherited ones.

Use these rules to specify conditions you want inherited down to the schema level. Some examples:

1. You want to prevent any table with "_dev" in its name from ever being monitored.
   1. Add a rule at the Warehouse level to Except tables where table name contains _dev
2. You want to ensure that tables without recent read or write activity on them are not monitored
   1. Add a rule at the Warehouse level to Monitor tables where there was activity in the last 30 days

We've added a rule under the Usage UI to let you select what tables to monitor based on tags. Under a Schema in the Usage UI, you can select Monitor tables where "tag" is one of: and specify a multi-select list of tags. This accompanies the rest of the rules we support, for a full list, see our documentation.

We've added two new rules to help you select which tables to monitor. Under the Schema level in the Usage UI there are two new rule types that can be added as include/exclude conditions:

1. Tables of a certain type (e.g. View, External, etc)
2. Last Activity was within 30 days

Tables of a certain type

This rule type can be used to make sure a certain type of asset is included or excluded from monitoring. Supported types depend on your warehouse and integration with Monte Carlo.

Last Activity was within 30 days

This rule can be used to ensure that only recenly active tables are monitored by Monte Carlo. "Recent Activity" is defined as any Read or Write activity.

Users of Data Explorer can now drill-down into a field profile metric in Data Explorer to see how that metric has trended over time. This helps a data analyst to validate an issue reported by a business partner. For example, if someone in marketing ops is reporting they’re suddenly seeing a bunch of nulls in a key field, it’s now much easier to come in and validate that spike in nulls without ever writing SQL or setting up a monitor.

We've made key improvements to the automatic threshold option in SQL Rules, giving users more accuracy and more control in detecting anomalies. The automatic threshold now supports:

• Low, medium, or high sensitivity. Users can adjust this through the UI or Monitors as Code, allowing for tighter or looser thresholds.
• Removing historical true positives from the machine learning, preserving the sensitivity of the thresholds. Incidents marked as fixed or helpful will be taken into account when training our ML model.
• Refactored model based on user feedback to improve accuracy.