ANOMALY_INSIGHTS!ADHOC_CALCULATE_ANOMALIES_FROM_CONFIG¶

Runs cost anomaly detection against a configuration without creating an anomaly monitor. Use this method to test a combination of tags and service types before you save the combination as a monitor.

Syntax¶

SNOWFLAKE.LOCAL.ANOMALY_INSIGHTS!ADHOC_CALCULATE_ANOMALIES_FROM_CONFIG(
  <config>,
  '<start_date>',
  '<end_date>' )

Arguments¶

config

Configuration that defines the scope to test. It uses the same keys as the configuration of a saved monitor, but it names each tag directly with the tagDatabase, tagSchema, tagName, and tagValues keys instead of taking a tag reference. For more information, see Monitor configuration.

Data type: VARIANT

'start_date'

Specifies the beginning of the time period for which consumption data is returned.

Data type: DATE

'end_date'

Specifies the end of the time period for which consumption data is returned.

Data type: DATE

Output¶

Returns a table with one row per day in the time period, with the following columns:

Column name	Data type	Description
USAGE_DATE	DATE	Day in UTC when the consumption occurred.
CONSUMPTION	NUMBER	Amount of consumption attributed to the monitor on this day.
FORECASTED_CONSUMPTION	NUMBER	Predicted consumption based on the anomaly-detecting algorithm.
CURRENCY_TYPE	VARCHAR	Unit of measure for the consumption, which corresponds to the monitor’s credit family.
LOWER_BOUND	NUMBER	Predicted lowest level of consumption based on the anomaly-detecting algorithm. Consumption levels below this value are considered anomalies.
UPPER_BOUND	NUMBER	Predicted highest level of consumption based on the anomaly-detecting algorithm. Consumption levels above this value are considered anomalies.
IS_ANOMALY	BOOLEAN	If `TRUE`, consumption fell outside the range defined by the lower and upper bounds, so Snowflake identified it as a cost anomaly.
ANOMALY_ID	VARCHAR	System-generated identifier for the anomaly. Empty when `IS_ANOMALY` is `FALSE`.
LAST_REFRESHED_AT	TIMESTAMP	Time when Snowflake last finished computing results for the monitor. This value is the same for every row.

Access control requirements¶

Users with any of the following roles can call this method:

ACCOUNTADMIN system role
GLOBALORGADMIN system role
SNOWFLAKE.APP_USAGE_ADMIN application role
SNOWFLAKE.APP_USAGE_VIEWER application role

Unlike ANOMALY_INSIGHTS!CREATE_MONITOR, this method doesn’t require the APPLYBUDGET privilege on the tags in the configuration.

Usage notes¶

This method names each tag directly rather than taking a tag reference, so you don’t call SYSTEM$REFERENCE to build the configuration. Nothing is saved and the configuration runs only one time, so no reference needs to be resolved.
The tag-name shape is the same one Snowflake returns when you read a configuration back, so you can pass a saved monitor’s configuration straight to this method.
The configuration must include at least one tag in resource_tags.tags or at least one entry in service_types. The method fails if it contains neither.
The results aren’t saved. To save a configuration as a monitor, use ANOMALY_INSIGHTS!CREATE_MONITOR, which takes tag references instead of tag names.
The method regenerates the full consumption time series and calculates any anomalies, so it takes longer to return than ANOMALY_INSIGHTS!GET_MONITOR_ANOMALIES.

Example¶

Test a configuration that tracks credits consumed by resources tagged with the department finance, along with all serverless task consumption in the account, for consumption between January 1, 2026, and March 31, 2026:

CALL SNOWFLAKE.LOCAL.ANOMALY_INSIGHTS!ADHOC_CALCULATE_ANOMALIES_FROM_CONFIG(
  PARSE_JSON('{
    "credit_family": "CREDITS",
    "resource_tags": {
      "operator": "UNION",
      "tags": [
        {
          "tagDatabase": "IT",
          "tagSchema": "WAREHOUSE_MANAGEMENT",
          "tagName": "DEPT",
          "tagValues": ["finance"]
        }
      ]
    },
    "service_types": ["SERVERLESS_TASK"]
  }'),
  '2026-01-01',
  '2026-03-31'
);