Master Data Quality Rules Catalog Worksheet
A data quality program starts with a written list of rules. This worksheet records each rule once with its field, check type, threshold, owner and frequency, so the team knows what is checked, by whom and how often.
| Rule ID | Domain | Field | Rule description | Check type | Pass threshold (%) | Owner | Run frequency | |
|---|---|---|---|---|---|---|---|---|
Add one row per rule. Keep rules testable: each should be answerable with a query that returns the failing records. Start with 15 to 25 rules that matter to planning, purchasing, costing and payments.
Turn rules into automated checks
Want these rules running against your ERP with exception lists sent to owners? Ask for an expert review.
Optional. The tool and every export work without this. No spam.
How to use the worksheet
Add a row per rule. Give it an ID with a domain prefix, name the field, and write the rule in plain words so a business user can read it. Pick the check type, set the pass threshold as a percent of records that must pass, and name an owner who will act on failures. Export to Excel and share with the governance group for review.
- One testable rule per row
- Name an owner who can fix failures
- Use 100% only where an exception is truly unacceptable
- Review the catalog quarterly
The five check types
Completeness asks whether a required value is present. Validity asks whether it conforms to a format, list or range. Uniqueness asks whether a real world thing appears once. Consistency asks whether related fields or records agree, for example an active BOM line pointing to an inactive item. Timeliness asks whether data is current enough to use, such as a review date within policy. These categories are common in data quality frameworks and keep the catalog balanced.
Choosing thresholds
Set thresholds from business impact, not habit. A missing tax ID on a payable vendor may justify 100%. A fuzzy duplicate rule will produce false positives, so a lower threshold with human review is more honest. Tighten thresholds gradually as the backlog of defects clears, and record changes in the catalog.
Common mistakes
Writing rules nobody can test, creating hundreds of rules before fixing the first ten, and having no owner. Rules that fail every day with no action teach people to ignore the report. Start small, publish results, and make failures visible to the owner.
Frequently Asked Questions
How many rules should I start with?
Fifteen to twenty-five that affect real decisions: planning, purchasing, costing, payments and compliance. A short catalog that is acted on beats a long one that is ignored.
Where do the rules run?
Rules can run as SQL or report queries in the ERP, as ETL checks in a data platform, or as validations at data entry. The catalog is the same regardless of where they run.
What is the difference between a rule and a KPI?
A rule tests individual records and produces a list of failures. A KPI summarizes pass rates across rules, for example percent complete. The scorecard worksheet turns rule results into KPIs.
Who approves new rules?
The data owner for the domain, with input from the steward. Changes should be dated and logged so trend reports stay comparable.
Once the rules are written, we can automate them as read-only queries against your ERP and send exception lists to owners.
Related Tools
ERP Data Quality Scorecard Worksheet
Enter record and pass counts per master data domain and the worksheet calculates quality percentages and an overall score.
ERP OperationsData Cleansing Defect Log Worksheet
Log each data quality defect with the affected records, severity, owner, effort and status, and see the total cleanup effort and cost.
ERP OperationsData Steward Role Description Template
A role description for the ERP master data steward: purpose, duties, authority, KPIs and escalation.
Costing & FinancePoor Master Data Quality Cost Calculator
Estimate the annual cost of master data errors from rework time, expediting and stewardship, and see how it scales with error rate.