A dataset is 99% accurate, 99% complete, 99% timely, but 30% non-unique. What's the dominant problem?

Uniqueness — duplicates inflate every aggregate and ruin joins

Your stock-level dataset is 100% accurate at midnight when it's snapshot, but your warehouse staff use it at 11 a.m. — by then 30% of items are sold. Which DQ dimension is silently failing?

Timeliness — freshness SLA doesn't match the use case

Six Dimensions of Data Quality — Semantic Web Academy

The six dimensions

The six axes you actually score

Accuracy — does the value reflect reality?
Completeness — are required fields present?
Consistency — does the same fact agree across systems?
Timeliness — is the value fresh enough for its use?
Uniqueness — exactly one row per real-world entity?
Validity — does the value satisfy its declared format / range?

A practical data-quality programme scores datasets on each axis, sets a target per axis per dataset (perfection is uneconomic), and instruments alerts on drift.

Vitals on a chart

Think of a hospital chart:

Accuracy = the right patient's vitals were entered.
Completeness = no blank allergy field.
Consistency = the same blood-pressure reading on the chart and the monitor.
Timeliness = today's vitals, not last week's.
Uniqueness = one chart per patient, not three with name variants.
Validity = the temperature is plausible (35–42 °C), not 410.

Bad scores on any one axis can kill a patient — same idea for a business decision.

Score one of your datasets

Pick a dataset you own and score it on the six dimensions.

›Which dimension is your lowest score — and is that the one your consumers are *actually* paying for?
›Where is a 'perfect' DQ target costing more than it saves (e.g. 100% accuracy on a field nobody reads)?
›What's the smallest contract you could publish next sprint that names a target per dimension for the top-3 most-used columns?

Reading in progress · 0 of 3 activities done