Overview

Monte Carlo's data exports make it easy to access and analyze critical metadata about your Data + AI Observability platform. It replaces our legacy "Insights" reports with a more aligned, flexible, and transparent approach to exporting Monte Carlo data.

With data exports, you can retrieve structured information on Monitors, Alerts, Events (the underlying events in an alert), and Assets—directly from the UI, via API/CLI, or through Snowflake data shares. These exports are refreshed daily and mirror what you see in Monte Carlo, allowing for easy integration into internal workflows, self-service reporting, and deep analysis.

For users seeking logs of each individual executions of a monitor, we recommend Monitor logs. This allow users to send detailed monitor execution logs to external systems via webhooks. Though it is a separate feature from data exports, the data is designed to be easily joined with data from Exports. Monitor logs are sent close to real-time and are useful for organizations that want to centralize auditing and compliance in their own tools or platforms.

A fifth data export, Consumption, is also available for customers on consumption pricing plans. It is available through the UI from the Settings > Billing page, via API/CLI, or through Snowflake data shares.

Monitor Runs, currently in private preview, exports the observed value and the threshold band it was evaluated against for each monitor execution. It is available via API/CLI or through Snowflake data shares, and joins to the other exports on the monitor ID.

Key Benefits

• Rich, structured data for Monitors, Alerts, Events, and Assets
• Reliable, daily-refreshed exports that match what you see in Monte Carlo
• Download directly from relevant Monte Carlo product pages for one-off analysis
• Snowflake private listing data share, configurable self-serve via the UI

Access methods

Note: The Alerts export includes both Alerts and their related Events.

CLI

Follow the CLI guide to get set up and see the export CLI docs for more information. Direct download and direct upload to S3 are both options.

API

Data exports are also available programmatically through the GraphQL API at https://api.getmontecarlo.com/graphql. See API Authentication to create a key and authenticate your requests.

Use the getDataExportUrl query to get a short-lived, presigned URL for the latest version of an export, then download the CSV from that URL — see DataExportURL in the API reference for the query and response shape. dataExportName accepts MONITORS, ALERTS, EVENTS, ASSETS, ALERTS_AND_EVENTS, MONITOR_LOGS, and CONSUMPTION.

Snowflake data share

For Snowflake users, we offer the ability to use data exports via Snowflake's Data Sharing capabilities. This is an opt-in feature.

Configuring via the UI

Account owners can enable and manage the Snowflake data share directly in Settings → Data sharing without contacting support.

To enable:

1. Click Enable data sharing.
2. Provide your Snowflake account details (shown in any Snowflake worksheet):
   • Organization — SELECT CURRENT_ORGANIZATION_NAME(); (e.g. ACMECORP)
   • Account name — SELECT CURRENT_ACCOUNT_NAME(); (e.g. ACME_PROD — the account name, not the account locator)
3. Click Enable data sharing to confirm.

Monte Carlo will add your account to the private listing — this can take a few hours to take effect.

To accept the listing in Snowflake:

1. In Snowsight, open Data sharing → External sharing.
2. Select the Shared with you tab.
3. Find the Monte Carlo Data Share listing and click Get to mount the database.

Once mounted, the shared database appears in your Snowflake account and is ready to query.

To update: click Reconfigure on the settings page.

To disable: click Disable on the private-listing share card. You can re-enable at any time.

Migrating from a legacy direct share

If your account uses a legacy direct share (locator-based), it appears as a separate card in Settings → Data sharing. You can migrate to the private listing at your own pace — both share types can run simultaneously, so consumers are never disrupted during the transition.

Option 1 — Run both in parallel (recommended)

1. Enable the private listing share as described above.
2. Have your consumers accept the new listing in Snowflake and switch their queries to the new database.
3. Once all consumers have migrated, disable the legacy share from its card.

Both shares expose the same tables and refresh on the same schedule — consumers can switch at their own pace.

Option 2 — Replace immediately

1. Enable the private listing share as described above.
2. Confirm your consumers are ready to switch, then disable the legacy share from its card.
3. Consumers accept the new private listing in Snowflake. When mounting the database, they can use the same name as the legacy share database — all existing queries and workflows will continue to work without any changes.

Note: Disabling a legacy share is irreversible and cannot be undone via the UI.

FAQs

Are exports filtered by domain or audience?

No. Exports include all relevant records. Users restricted to specific domains do not have access to data exports.

Where can I find the export option in the UI?

Look for export/download buttons within the Monitors, Alerts, and Assets pages. There is no centralized "Exports" tab at this time.

How often are exports refreshed?

Daily. Report generation generally finishes around 11:00 UTC, though there may be some day-to-day variation.

Data export schemas

Each export contains a clean, well-documented schema. Fields are consistently named, and join keys are aligned across exports.

Note: All columns are delivered as TEXT in the data share regardless of the semantic type shown below. Cast to the appropriate type on read (e.g. TO_TIMESTAMP_NTZ(created_at)).

Monitors

For Table Monitors, the ASSETS and ASSET_IDS fields include up to 1,000 assets per monitor; larger monitors are truncated to the first 1,000. Tip: when syncing exports to Snowflake via Data share, use parse_json(assets) (Snowflake) to query individual assets.

CONNECTION_NAME is not included for table monitors, query performance monitors

Alerts

Events

Monitor Runs

Note: This feature is in private preview — reach out to your account representative to request access.

For each monitor execution, the observed value and the threshold band it was evaluated against — one row per time series per run. Only time series with a modeled threshold (an active rule) are included; context-only metrics with no threshold are omitted.

Access: Available via the Snowflake data share, the montecarlo export CLI, and the GraphQL API. It is not available as a UI download.

History: Via Snowflake Data Share, history accrues from when the export is enabled, up to a rolling 90 days. Via API/CLI, the last 2 days of runs are included.

Both static (manual GT/LT/EQ/…) and dynamic (AUTO) thresholds are represented.

Assets

Consumption