Add scorecard items
Record each domain, system, product, dataset, or critical element with its owner and relative weight.
Data quality management tool
Build a practical data quality scorecard across your portfolio
Assess domains, systems, data products, datasets, or critical data elements using seven quality dimensions, weighted priorities, thresholds, trends, ownership, issue counts, and accountable actions.
Transparent and privacy-consciousThe method is deterministic, the formula is shown, and no data is sent to external services. Results depend on the information you provide.
How it works
Create a repeatable scorecard that connects measurement to ownership and action.
Score and compare
Enter seven dimension scores, a target threshold, prior-period score, open issues, and the next action.
Prioritise remediation
Review weighted results, RAG status, trend, target gap, weakest dimensions, and prioritised actions.
Methodology, limitations, and practical use
Use the scorecard as a governance and prioritisation aid, not as a substitute for source-system profiling, statistical validation, audit evidence, or professional judgement.
Use consistent evidence
Define each dimension, sampling period, population, tolerance, and evidence source before scoring. Keep the definition stable across reporting periods.
Interpret trends carefully
A trend may reflect a real control change, a different sample, a revised rule, or improved issue detection. Record the reason for material movement.
Close actions with evidence
Assign one owner, a due date, expected outcome, and validation evidence. Re-score only after the remediation has been tested.
Frequently asked questions
What does the portfolio score represent?
It is the weighted average of item-level scores. Each item score is the arithmetic mean of seven data quality dimensions, and each item contributes according to its relative weight.
How should we choose weights?
Use a consistent relative scale based on business criticality, regulatory impact, customer impact, financial exposure, or operational dependency. A critical dataset might receive a weight of 5 while a lower-impact dataset receives 1.
Do weights need to add up to 100?
No. The calculation divides by the total weight, so any positive and consistent relative scale works.
How is RAG status determined?
Green means the score meets or exceeds its target. Amber means it is below target by no more than 10 points. Red means it is more than 10 points below target.
How is trend calculated?
The tool compares the current item score with the entered prior-period score. Movement greater than 0.4 points is improving, below −0.4 is declining, and movement within that range is stable.
Can we score critical data elements and systems together?
Yes, but interpret the portfolio carefully. Mixed item types can be useful for executive oversight, while operational teams may prefer separate scorecards by type for more comparable results.
What evidence should support a dimension score?
Use documented profiling rules, reconciliations, validation tests, control logs, issue registers, source-to-target checks, service-level measurements, and representative samples.
Why is duplicate-name checking included?
Duplicate names can cause ambiguous ownership, double counting, and export problems. Use distinct names or qualify them with the system, domain, region, or product.
What does a zero issue count mean?
It may mean no issues are open, but it may also indicate incomplete issue capture. Confirm that monitoring, triage, and logging processes are operating before treating zero as assurance.
Can the CSV import contain formulas?
No. Import plain values only. The expected columns are Type, Name, Owner, the seven dimension scores, Weight, Target, Prior, Issues, and Action.
Is data sent outside the website?
No external API is used. Browser-based import and export stay local. Form submission is processed by the page itself unless the site owner later adds secure storage.
Is this scorecard an audit or certification?
No. It is a management aid based on user-supplied information. It does not independently verify controls, evidence, regulatory compliance, or certification readiness.