Administer Data Quality Settings¶

Alation Cloud Service Applies to Alation Cloud Service instances of Alation

The Admin Settings page is the central control panel for Alation Data Quality administrators. It provides access to manage global settings for your organization. From this page, you can enable or disable features, configure role-based permissions for users, set up approval workflows for monitors, and manage webhook integrations.

Access to this page is controlled by user role and permissions so that you can tailor the Data Quality module to fit your organization’s governance policies and operational needs.

Permission Management¶

The following table shows the actions users can perform within the Data Quality module based on their assigned Alation role. The Can view permission is mandatory for all roles. You can grant or revoke write access by selecting the Can create monitor checkbox for each role.

Role	Create & Manage Monitors	View DQ Results
Catalog Admin	Yes (default)	Yes (default)
Server Admin	Yes (default)	Yes (default)
Source Admin	Yes (default)	Yes (default)
Steward	Yes (default)	Yes (default)
Composer	View only (default)	Yes (default)
Explorer	View only (default)	Yes (default)
Viewer	View only (default)	Yes (default)

Configure Admin Settings¶

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
Configure the Data Quality capabilities:
- Data Profiling: Enables on-demand statistical analysis for tables and columns, such as row counts, null rates, and distributions. Disable this feature to restrict compute-intensive profiling operations.
- Data Sampling: Enables queries to run on a sample of rows instead of a full table during profiling or failed row analysis. This reduces the compute load on large datasets.
  
  If enabled, you can also select Sample using service account to execute sampling queries with a designated service account.
- Export CSV: Controls the ability for users to export Data Quality results, such as check outcomes or failed row data, to a CSV file. Disable this for organizations with strict data export policies.
- Recommend Checks: Activates the LLM-powered engine that suggests relevant DQ checks during monitor creation based on online column metadata. Disable this if your organization restricts LLM-assisted features.
In the permissions management section, you can control the actions users can perform based on their assigned Alation role.
- To grant write access, select the Can create monitor checkbox for each role. (Default: Can view)
In the approval workflow setup section, you must configure the approval process for a DQ monitor:
- Enable the Workflow: Enables the approval workflow. When enabled, users must submit new monitors or changes for review before they become active.
- Approver Group: Select the Alation users or groups authorized to review and approve monitor submissions. Only users in this group receive approval notifications.
- Admin Bypass: Designate specific admin users who can publish monitors immediately, bypassing the approval process.
In the webhook configuration section, configure the outbound webhooks to send real-time notifications for Data Quality events.
- Registration: Register a new webhook by providing a Display name and the endpoint URL.
- Notification and incidents scope: Defines the granularity of the notifications sent by the webhook.
  - Per monitor base: Select to send one notification that summarizes the status of an entire monitor run.
  - Per check base: Select to send a separate notification for each individual check within a monitor run.
- Authentication mechanisms: Defines the authentication method for the webhook subscription.
  Static Header Token: Select to authenticate using a fixed token sent in an HTTP header.
  
  URL: Enter the destination URL of the external system that receives the webhook notifications.
  
  Token Header: Enter the name of the HTTP header that contains the authentication token.
  
  Token: Enter the secret token value that the external system uses to verify the request.
In the top-right corner, click Save.

Check Recommendation Skills¶

Check recommendation skills guide the AI-powered check recommendation engine. Each skill defines patterns, rules, and best practices that the engine uses to suggest relevant data quality checks for your columns. Alation provides universal skills that cover common data quality scenarios. You can also create custom skills tailored to your organization’s specific requirements.

There are two types of skills:

Universal skills: Built-in skills provided by Alation that address common data quality patterns. These skills are read-only and cannot be modified or deleted.
Custom skills: Skills you create to address your organization’s unique data quality requirements. You can edit or delete custom skills at any time.

View Skills¶

To view existing check recommendation skills:

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
In the left sidebar, click Check Rec. Skills.
The skills list displays all universal and custom skills. Click a skill card to view its details.

Note

Universal skills display in read-only mode. You can view the skill content but cannot modify it.

Create a Custom Skill¶

You can create custom skills in two ways: by using the copilot chat to generate skill content with AI assistance, or by uploading a skill file directly.

Create a Skill with Copilot Chat¶

The copilot chat provides an AI-assisted workflow for generating skill content. You describe the data quality pattern or rule you want to enforce, and the copilot generates the skill content for you.

To create a skill using the copilot chat:

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
In the left sidebar, click Check Rec. Skills.
Click Create Skill.
In the Chat with Copilot tab, describe the data quality pattern or rule you want the skill to address. For example, you might describe a naming convention, a validation rule, or a data pattern specific to your organization.
The copilot generates skill content based on your description. Review the generated content in the skill editor on the left side of the dialog.
Edit the skill content as needed to refine the instructions.
Enter a Skill Name and Description for your skill.
Click Save Skill.

Create a Skill by Uploading a File¶

You can create a skill by uploading a text or markdown file that contains the skill instructions.

To create a skill by uploading a file:

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
In the left sidebar, click Check Rec. Skills.
Click Create Skill.
Click the Upload a Skill tab.
Click Upload and select a file containing your skill instructions.
The file content appears in the skill editor. Review and edit the content as needed.
Enter a Skill Name and Description for your skill.
Click Save Skill.

Edit a Custom Skill¶

You can modify custom skills to update their instructions or metadata.

To edit a custom skill:

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
In the left sidebar, click Check Rec. Skills.
Locate the skill you want to edit and click Edit.
Update the skill content, name, or description as needed.
Click Save Skill.

Delete a Custom Skill¶

You can delete custom skills that are no longer needed. Deleting a skill removes it from the check recommendation engine.

To delete a custom skill:

Navigate to the Data Quality dashboard.
In the top-right corner, click Admin Settings.
In the left sidebar, click Check Rec. Skills.
Locate the skill you want to delete and click Delete.
In the confirmation dialog, click Delete to confirm.

Warning

Deleting a skill cannot be undone. Once deleted, the skill is permanently removed and will no longer influence check recommendations.