Master Data Quality Dimensions: Data Quality Dimensions are not just frameworks—they are the heartbeat of enterprise decision-making. According to Gartner, poor data quality costs organizations an average of $12.9 million annually, while IBM reports that bad data affects over 30% of business revenues. Despite these facts, only 16% of organizations have a mature data quality program in place.

This article breaks down the five critical dimensions of data quality, explains how they’re implemented, and demonstrates how ServiceNow and Data Trust tools can elevate your organization’s data maturity through metadata-driven rules and conditional logic.

---

🧠 Why It Matters

• Decision-making depends on data accuracy
• AI models rely on consistent and clean data
• Regulations like GDPR and HIPAA demand traceability and data lineage
• Enterprise automation fails without timely and complete data

Now, let’s examine each dimension in depth and how they’re put into action.

---

✅ Five Dimensions of Data Quality

Dimension	Description	Key Components
Accuracy	Reflects the real-world truth	Reference validation, correct values, business rule checks
Completeness	Measures how much data is present	Required fields, fill rates, null checks
Consistency	Uniformity across systems	Referential integrity, business logic alignment
Timeliness	Data’s freshness and update recency	Timestamps, sync intervals, latency
Uniqueness	Ensures no duplicates exist	Key constraints, match rules, deduplication

---

🧱 Constructing the Dimensions

Each dimension is powered by:

• Rules & Thresholds (e.g., Accuracy ≥ 95%)
• Query Logic (SQL or GlideRecord)
• Metadata Profiles (column type, requirement level)
• Ownership Models (defined stewards and systems of record)

These pieces ensure not only quality measurement—but also actionability.

---

🔍 Building the Queries: Conditions, Checks, Subselects

Standard SQL Subselect Pattern:

sqlCopyEditSELECT col1, col2
FROM table
WHERE condition
AND NOT EXISTS (
  SELECT 1 FROM other_table WHERE table.id = other_table.ref_id
)

Example: Null Value Check for Completeness

sqlCopyEditSELECT employee_id, last_name
FROM hr_profile
WHERE last_name IS NULL;

GlideRecord in ServiceNow:

javascriptCopyEditvar gr = new GlideRecord('cmdb_ci');
gr.addNullQuery('serial_number');
gr.query();
while (gr.next()) {
  gs.info(gr.name);
}

---

🧾 Using Metadata and Column Queries

Purpose	Metadata Used
Identify Required Fields	sys_dictionary.mandatory=true
Validate Data Types	element.column_type=string/date
Track Data Freshness	sys_audit.timestamp
Define Relationships	sys_db_object, sys_dictionary

Sample Metadata Query:

sqlCopyEditSELECT name, column_label, mandatory
FROM sys_dictionary
WHERE name = 'incident';

---

⚙️ Conditional Builder in ServiceNow

The Conditional Builder empowers users to build complex logic with ease—no code needed.

Key Features:

• Drag-and-drop conditions
• Field/operator/value pairing
• Logical AND/OR combinations
• Real-time rule preview

Used For:

• SLA start/stop rules
• Incident resolution criteria
• Data validation checks
• Metadata-triggered automation

---

🔐 Metadata in Data Trust

In ServiceNow’s Data Trust, metadata governs:

Metadata Type	Usage
Column Attributes	Data validation and expectations
Update History	Track freshness and audit trail
Ownership Annotations	Assign accountability
Trust Scores	Evaluate field reliability over time

This metadata isn’t passive—it actively powers governance, quality scoring, and confidence reporting.

---

📚 Where to Learn More

• ServiceNow Docs: Conditional Builder
• Data Management Association (DAMA): DMBOK2 Framework
• MIT CDOIQ Symposium: Best practices on metadata and data trust

Other Master Data Quality Dimensions

• 6 Data Governance Principles for Reports and Dashboards
• AI Demands: Data Stewards
• A-Z Data Fabric Glossary
• Artificial Intelligence A-Z Glossary
• Business Process Improvement Glossary
• Comprehensive Guide to Data Stewardship
• DAMA Data Management Guide
• Data Quality Management With Databricks | Databricks
• Data Quality: Why It Matters and How to Achieve It
• Data Stewards
• Understanding the Impact of Bad Data – DATAVERSITY