Group by

Last modified on 26-Apr-24

Use a group by configuration to collect and present check results by category.

checks for fact_internet_sales:
  - group by: # Not supported in Soda Core
      query: |
        SELECT sales_territory_key, AVG(discount_amount) as average_discount
        FROM fact_internet_sales
        GROUP BY sales_territory_key
      fields:
        - sales_territory_key
      checks:
        - average_discount:
            fail: when > 40
            name: Average discount percentage is less than 40% (grouped by sales territory)

✖️    Requires Soda Core Scientific (included in a Soda Agent)
✖️    Supported in Soda Core
✔️    Supported in Soda Library + Soda Cloud
✔️    Supported in Soda Cloud Agreements + Soda Agent
✔️    Supported by SodaGPT
✖️    Available as a no-code check

Define a group by configuration checks
Group by check results
Optional check configurations
    Add multiple group configurations
    Change configurations and preserve check history
    Track anomalies and relative changes by group
Troubleshoot
Go further

Define a group by configuration

For an individual dataset, add a group by configuration to specify the categories into which Soda must group the check results.

The example below uses a SQL query to define a custom metric for the fact_internet_sales dataset. It calculates the average order discount based on the contents of the discount_amount column, then groups the results according to the value in the sales_territory_key. This check supports up to a maximum of 1000 groups.

The check itself uses the custom metric average_discount and an alert configuration to determine if the measurement for each group passes, warns, or fails. In this case, any calculated measurement for average that exceeds 40 results in a fail.

checks for fact_internet_sales:
  - group by:
      group_limit: 10
      name: average discount
      query: |
        SELECT sales_territory_key, AVG(discount_amount) as average_discount
        FROM fact_internet_sales
        GROUP BY sales_territory_key
      fields:
        - sales_territory_key
      checks:
        - average_discount:
            fail: when > 40
            name: Average discount percentage is less than 40% (grouped-by sales territory)

`group by`	required	configuration section label
`group_limit`	optional	the maximum number of groups, or column values, into which Soda must categorize the results. This value must correspond with the number of unique values in the column you identify in the `fields` section; see example below. This check supports up to a maximum of 1000 groups.
`group_name`	optional	specify a name for the group; Soda does not evaluate this parameter
`query`	required	custom query subsection label. The nested SQL query defines the custom metric `average_discount`
`fields`	required	column subsection label
`sales_territory_key`	required	column identifier; the values in this column identify how Soda groups the results.
`checks`	required	check subsection label
`average_discount`	required	custom metric identifier
`fail: when > 40`	required	fail condition and threshold
`warn: when between 50 and 60`	only one alert condition is required	warn condition and threshold
`name`	optional	custom name for the check; it not defined, Soda derives the name of the check in Soda Cloud from the check syntax

You can also use multi-column groups in a group by check, as in the example below that groups results both by gender and english_education.

checks for dim_customer:
    - group by:
        query: |
            SELECT
                gender,
                english_education,
                sum(total_children) as sum_total_children
            FROM dim_customer
            GROUP BY gender, english_education
        fields:
            - gender
            - english_education
        checks:
            - sum_total_children:
                fail: when < 100000
                name: Total number of children

Group by check results

When you run a scan that includes checks nested in a group by configuration, the output in Soda Library CLI groups the results according to the unique values in the column you identified in the fields subsection. The number of unique values in the column must match the value you provided for group_limit.

In the example results below, the calculated average of average_discount for each sales territory is less than 40%, so the check for each group passed. The value in the square brackets next to the custom check name identifies the group which, in this case, is a number that corresponds to a territory.

Soda 1.0.x
Soda Core 3.0.x
Scan summary:
11/11 checks PASSED: 
    fact_internet_sales in adventureworks
      group by [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [8] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [10] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [9] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [7] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [1] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [5] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [2] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [4] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [6] [PASSED]
      Average discount percentage is less than 40% (grouped-by sales territory) [3] [PASSED]
All is good. No failures. No warnings. No errors.
Sending results to Soda Cloud
Soda Cloud Trace: 14733***37

In Soda Cloud, the check results appear by territory.

group-by-4

Be aware that a check that contains one or more alert configurations only ever yields a single check result; one check yields one check result. If your check triggers both a warn and a fail, the check result only displays the more severe, failed check result. (Schema checks behave slightly differently; see Schema checks.)

Using the following example, Soda Library, during a scan, discovers that the data in the dataset triggers both alerts, but the check result is still Only 1 warning. Nonetheless, the results in the CLI still display both alerts as having both triggered a [WARNED] state.

checks for dim_customer:
  - row_count:
      warn:
        when > 2
        when < 0

Soda Library 1.0.x
Soda Core 3.0.x
Scan summary:
1/1 check WARNED: 
    dim_customer in adventureworks
      row_count warn when > 2 when > 3 [WARNED]
        check_value: 18484
Only 1 warning. 0 failure. 0 errors. 0 pass.
Sending results to Soda Cloud
Soda Cloud Trace: 42812***

The check in the example below data triggers both warn alerts and the fail alert, but only returns a single check result, the more severe Oops! 1 failures.

checks for dim_product:
  - sum(safety_stock_level):
      name: Stock levels are safe
      warn:
        when > 0
      fail:
        when > 0

Soda Library 1.0.x
Soda Core 3.0.x
Scan summary:
1/1 check FAILED: 
    dim_product in adventureworks
      Stock levels are safe [FAILED]
        check_value: 275936
Oops! 1 failures. 0 warnings. 0 errors. 0 pass.
Sending results to Soda Cloud
Soda Cloud Trace: 6016***

Optional check configurations

Supported	Configuration	Documentation
✓	Define a name for a group by; see example.	Customize check names
✓	Add an identity to a check.	Add a check identity
✓	Define alert configurations to specify warn and fail alert conditions; see example	Add alert configurations
	Apply an in-check filter to return results for a specific portion of the data in your dataset.	-
✓	Use quotes when identifying dataset or column names; see example. Note that the type of quotes you use must match that which your data source uses. For example, BigQuery uses a backtick (`) as a quotation mark.	Use quotes in a check
✓	Use wildcard characters in the value in the check.	Use wildcard values as you would with SQL.
	Use for each to apply group by checks to multiple datasets in one scan.	-
	Apply a dataset filter to partition data during a scan.	-

Example with check name

When the check results appear in Soda Cloud, the checks use the name you define for the check or, if you do not specify a name parameter, the syntax of the check itself.

checks for dim_employee:
  - group by:
      name: Grouped vacation hours
      group_limit: 2
      query: |
        SELECT marital_status, AVG(vacation_hours) as vacation_hours
        FROM dim_employee
        GROUP BY marital_status
      fields:
        - marital_status
      checks:
        - vacation_hours > 60:
            name: Too many vacation hours

Example with identity

checks for dim_employee:
  - group by:
    ...
      checks:
        - vacation_hours > 0:
            identity: custom_identity

Example with alert configuration

Be aware that Soda only ever returns a single check result per check. See Expect one check result for details.

checks for dim_employee:
  - group by:
      group_limit: 2
      query: |
        SELECT marital_status, AVG(vacation_hours) as vacation_hours
        FROM dim_employee
        GROUP BY marital_status
      fields:
        - marital_status
      checks:
        - vacation_hours:
            fail: when > 65
            warn: when between 50 and 65
            name: Too many vacation hours

Example with quotes

Note that the type of quotes you use must match that which your data source uses. For example, BigQuery uses a backtick (`) as a quotation mark.

checks for dim_employee:
  - group by:
      group_limit: 2
      query: |
        SELECT "marital_status", AVG("vacation_hours") as vacation_hours
        FROM "dim_employee"
        GROUP BY marital_status
      fields:
        - marital_status
      checks:
        - vacation_hours > 60:
            name: Too many vacation hours

Add multiple group configurations

You can add multiple group by configurations to the same dataset in the same checks YAML file and produce separately grouped check results. To do so, you must include an identifier in the group by configuration key of the extra group configurations you add, as in the example below.

The first group by in the example requires no identifier, though for readability and completeness, you have the option of adding one.
The second group by includes title as an identifier to differentiate it from the first.

checks for dim_employee:
  - group by:
      query: |
        SELECT marital_status, AVG(vacation_hours) as vacation_hours, MAX(vacation_hours) as max_vacation_hours
        FROM dim_employee
        GROUP BY marital_status
      fields:
        - marital_status
      checks:
        - vacation_hours > 0
        - max_vacation_hours < 100:
            name: MAX vacation hours less than 100 [marital_status]
  - group by title:
      query: |
        SELECT title, AVG(vacation_hours) as vacation_hours, MAX(vacation_hours) as max_vacation_hours
        FROM dim_employee
        GROUP BY title
      fields:
        - title
      checks:
        - vacation_hours > 0:
        - max_vacation_hours < 100:
            name: MAX vacation hours less than 100 [title]

When the check results appear in Soda Cloud, the checks use the name you define for each check or, if you do not specify a name parameter, the syntax of the checks themselves. The group by identifiers appear in the CLI output, but do not appear the the check results in Soda Cloud.

Soda Library 1.3.1
Soda Core 3.0.47
By downloading and using Soda Library, you agree to Sodas Terms & Conditions (https://go.soda.io/t&c) and Privacy Policy (https://go.soda.io/privacy). 
Scan summary:
138/140 checks PASSED: 
    dim_employee in adventureworks
      group by [PASSED]
      vacation_hours > 0 [M] [PASSED]
      MAX vacation hours less than 100 [marital_status] [M] [PASSED]
      vacation_hours > 0 [S] [PASSED]
      MAX vacation hours less than 100 [marital_status] [S] [PASSED]
      vacation_hours > 0 [Database Administrator] [PASSED]
      MAX vacation hours less than 100 [title] [Database Administrator] [PASSED]
      vacation_hours > 0 [Design Engineer] [PASSED]
      MAX vacation hours less than 100 [title] [Design Engineer] [PASSED]
      vacation_hours > 0 [Production Supervisor - WC20] [PASSED]
      MAX vacation hours less than 100 [title] [Production Supervisor - WC20] [PASSED]
      vacation_hours > 0 [Research and Development Engineer] [PASSED]
      MAX vacation hours less than 100 [title] [Research and Development Engineer] [PASSED]
      vacation_hours > 0 [Research and Development Manager] [PASSED]
      ...
2/140 checks FAILED: 
    dim_employee in adventureworks
      group by title [FAILED]
      vacation_hours > 0 [Chief Financial Officer] [FAILED]
        check_value: 0E-20
Oops! 2 failures. 0 warnings. 0 errors. 138 pass.
Sending results to Soda Cloud
Soda Cloud Trace: 693****

multiple-group-results

group1-results

group2-results

group3-results

group4-results

Change configurations and preserve check history

When you make changes to your group by configuration, some changes result in a resetting of the check history in Soda Cloud.

The following changes result in a reset of a group by check’s history; all historical measurements disappear.

change the list of fields, either adding, removing, or changing an existing field
change the group by identifier when more than one group by configurations exist; see Add multiple group configurations
change the syntax of a group by check such as, a change to the threshold value; see Configure variables in SodaCL to mitigate disruption using dynamic threshold values

The following changes result in no changes to the check’s history; Soda preserves all historical measurements to a maximum of 90 days.

change the SQL query that forms part of the group by configuration
add a check to the group by configuration

Further, you can add an identity parameter to a group by check to be able to make changes to the check and still preserve its historical measurements, as in the example below. See also: Add a check identity.

checks for dim_employee:
  - group by:
    ...
      checks:
        - vacation_hours > 0:
            identity: custom_identity

Track anomalies and relative changes by group

You can use a group by configuration to detect anomalies by category, and monitor relative changes over time in each category.

✔️    Requires Soda Core Scientific for anomaly check (included in a Soda Agent)
✖️    Supported in Soda Core
✔️    Supported in Soda Library 1.1.27 or greater + Soda Cloud
✔️    Supported in Soda Cloud Agreements + Soda Agent 0.8.57 or greater
✖️    Supported by SodaGPT
✖️    Available as a no-code check

The following example includes three checks grouped by gender.

The first check uses the custom metric average_children to collect measurements and gauge them against an absolute threshold of 2.
Soda Cloud displays the check results grouped by gender.
The second check uses the same custom metric to detect anomalous measurements relative to previous measurements. Soda must collect a minimum of four, regular-cadence, measurements to have enough data from which to gauge an anomolous measurement. Until it has enough measurements, Soda returns a check result of [NOT EVALUATED].
Soda Cloud displays any detected anomalies grouped by gender.
The third check uses the same custom metric to detect changes over time in the calculated average measurement, and gauge the measurement against a threshold of between -5 and 5 relative to the previously-recorded measurement. See Change-over-time thresholds for supported syntax variations for change-over-time checks.
Soda Cloud displays any detected changes grouped by gender.

checks for dim_customer:
  - group by:
      name: Group by gender
      query: |
        SELECT gender, AVG(total_children) as average_children
        FROM dim_customer
        GROUP BY gender
      fields:
        - gender
      checks:
        - average_children > 2:
            name: Average children per gender should be more than 2
        - anomaly detection for average_children:
            name: Detect anomaly for average children
        - change for average_children between -5 and 5:
            name: Detect unexpected changes for average children

group-anomaly

Troubleshoot

If you use an optional group_limit parameter, you must always match the number of unique values in the group by column to the value you provide for the parameter.

In the following example, the number of unique values in the sales_territory_key column is greater than the group_limit: 2 so Soda does not evaluate the check.

checks for dim_employee:
  - group by:
      group_limit: 2
      query: |
        SELECT sales_territory_key, AVG(vacation_hours) as vacation_calc
        FROM dim_employee
        GROUP BY sales_territory_key
      fields:
        - sales_territory_key
      checks:
        - vacation_calc > 60:
            name: Reasonable vacation hours

Soda Library 1.0.x
Soda Core 3.0.x
Evaluation of check group by failed: Total number of groups 11 exceeds configured group limit: 2
  | Total number of groups 11 exceeds configured group limit: 2
  +-> line=2,col=5 in checks_groupby.yml
Scan summary:
1/1 check NOT EVALUATED: 
    dim_employee in adventureworks
      group by [NOT EVALUATED]
1 checks not evaluated.
1 errors.
Oops! 1 error. 0 failures. 0 warnings. 0 pass.
ERRORS:
Evaluation of check group by failed: Total number of groups 11 exceeds configured group limit: 2
  | Total number of groups 11 exceeds configured group limit: 2
  +-> line=2,col=5 in checks_groupby.yml

Resolve the issue by increasing the group limit value, or by removing the parameter entirely.

checks for dim_employee:
  - group by:
      group_limit: 11
      query: |
        SELECT sales_territory_key, AVG(vacation_hours) as vacation_calc
        FROM dim_employee
        GROUP BY sales_territory_key
      fields:
        - sales_territory_key
      checks:
        - vacation_calc > 20:
            name: Too much vacation

Soda Library 1.0.x
Soda Core 3.0.x
Scan summary:
12/12 checks FAILED: 
    dim_employee in adventureworks
      group by [FAILED]
      Too much vacation [3] [FAILED]
        check_value: 24.0000000000000000
      Too much vacation [8] [FAILED]
        check_value: 35.0000000000000000
      Too much vacation [11] [FAILED]
        check_value: 51.3297872340425532
      Too much vacation [9] [FAILED]
        check_value: 36.0000000000000000
      Too much vacation [7] [FAILED]
        check_value: 34.0000000000000000
      Too much vacation [10] [FAILED]
        check_value: 37.0000000000000000
      Too much vacation [1] [FAILED]
        check_value: 28.0000000000000000
      Too much vacation [5] [FAILED]
        check_value: 29.0000000000000000
      Too much vacation [4] [FAILED]
        check_value: 26.5000000000000000
      Too much vacation [2] [FAILED]
        check_value: 38.0000000000000000
      Too much vacation [6] [FAILED]
        check_value: 32.0000000000000000
Oops! 12 failures. 0 warnings. 0 errors. 0 pass.

Go further

Use a group evolution check to surface changes in groups in a dataset.
Learn more about alert configurations.
Learn more about SodaCL metrics and checks in general.
Need help? Join the Soda community on Slack.

Was this documentation helpful?

What could we do to improve this page?

Suggest a docs change in GitHub.
Share feedback in the Soda community on Slack.

Documentation always applies to the latest version of Soda products
Last modified on 26-Apr-24