Data Governance
Overview
Effective governance is essential to maintaining a reliable and trustworthy Configuration Management Database (CMDB). Without proper governance, CMDB data can quickly become incomplete, inconsistent, or outdated leading to inaccurate reporting, failed automations, and poor operational decisions.
ServiceNow provides built-in capabilities to monitor and enforce data quality through CMDB Health, which evaluates the CMDB across three key dimensions:
- Completeness – Are required and recommended fields populated?
- Compliance – Do CIs meet defined standards and policies?
- Correctness – Is the data accurate, current, and properly related?
Together, these metrics ensure that CMDB data is not only present, but also usable and reliable for IT operations.
CMDB Health Scorecards
The Completeness scorecard measures how well Configuration Items (CIs) are populated with required and recommended data.
Key Metrics
- Required Fields: Defined by setting the Mandatory attribute at the dictionary level; measures the percentage of CIs missing required data.
- Recommended Fields: Defined in the CI Class Manager; measures the percentage of CIs missing suggested (but not enforced) data.
Governance Considerations
Completeness ensures that critical operational data is available for incident and change routing, ownership and accountability, and asset tracking and reporting.
Key Questions
- Are all required fields populated for each CI?
- Are important non-discoverable fields (for example, ownership and support group) filled in?
Data Governance Processes
Duplicate Remediation
Duplicate records significantly impact CMDB reliability and reporting. ServiceNow provides two primary methods to remediate these duplicates:
Duplicate CI Remediation This guided process assists in resolving duplicates by offering options to merge, delete, or retain records. The workflow involves reviewing duplicate CIs, comparing their attributes, and selecting the best action to maintain data integrity. A Main CI is designated based on specific criteria: the most recently updated record, the record with the most existing relationships, or the oldest record (created first).
De-duplication Templates Templates enable bulk resolution by defining identification criteria and automated actions, such as merging or deleting. These can be assigned to open de-duplication tasks or published to run on a scheduled basis, providing an efficient, automated way to manage duplicates at scale.
Reclassification
Reclassification is the process of changing the class of a Configuration Item (CI) when it has been misclassified or when new information becomes available.
Types of Reclassification:
Upgrade
- Moves a CI to a more specific (child) class
- Example:
Server → Windows Server
Downgrade
- Moves a CI to a more general (parent) class
- May result in loss of attributes not present in the parent class
- Example:
Windows Server → Server
Switch
- Moves a CI to a different branch in the class hierarchy
- May result in attribute data loss
- Example:
Linux Server → Windows Server
Global System Properties
These properties control whether automatic reclassification occurs:
glide.class.upgrade.enabledglide.class.downgrade.enabledglide.class.switch.enabled
Setting these to false:
- Prevents automatic reclassification
- Generates reclassification tasks instead
You can further refine behavior using these properties to allow data updates while specifically blocking the class change:
glide.identification_engine.update_without_switch_enabledglide.identification_engine.update_without_downgrade_enabledglide.identification_engine.update_without_upgrade_enabled
When set to true:
- CI attributes are updated
- Reclassification is skipped
Reclassification Restriction Rules
Reclassification restriction rules allow control at the class level, rather than globally.
- Restrict downgrade and switch actions between specific classes
- Prevent unwanted data loss during reclassification
- Allow attribute updates while blocking class changes
Summary
CMDB data governance is centered around ensuring data quality, consistency, and reliability. By leveraging CMDB Health and supporting processes:
- Completeness ensures required data is populated
- Compliance ensures adherence to standards
- Correctness ensures data accuracy and integrity
Supporting processes such as duplicate remediation and reclassification controls further strengthen governance by maintaining clean, accurate, and well-structured data.
A mature CMDB requires continuous monitoring and enforcement of these principles to remain a trusted source of truth for the organization.