The Six Dimensions of Data Quality

Accuracy

Does data correctly represent the real-world entity? Example: Customer address matches their actual address. Verify against source of truth.

Completeness

Are all expected records and fields present? Example: Missing customer emails in a 95% complete dataset. Measure: NULL percentage per column.

Consistency

Is data the same across systems? Example: Revenue differs between finance and analytics dashboards. Reconcile across systems.

Timeliness

Is data available when needed? Example: Dashboard shows yesterday's data instead of today's. Measure: Data freshness SLA.

Validity

Does data conform to defined formats/constraints? Example: Email field contains valid email format. Enforce with schema validation.

Uniqueness

Are there duplicate records? Example: Same order appearing twice in the warehouse. Detect with COUNT(DISTINCT) vs COUNT(*).

Code Example

Data quality checks with SQL

-- Completeness check
SELECT
COUNT() AS total_rows,
COUNT(email) AS non_null_emails,
ROUND(COUNT(email) * 100.0 / COUNT(), 2) AS completeness_pct
FROM users;

-- Uniqueness check
SELECT
order_id,
COUNT() AS occurrences
FROM orders
GROUP BY order_id
HAVING COUNT() > 1; -- Duplicates!

-- Validity check
SELECT *
FROM users
WHERE email NOT LIKE '%@%.%'; -- Invalid emails

-- Timeness check
SELECT
MAX(updated_at) AS last_update,
EXTRACT(EPOCH FROM (NOW() - MAX(updated_at))) / 3600 AS hours_since_update
FROM fact_orders;

Best Practices

• Define quality dimensions for each dataset
• Implement automated checks in pipelines
• Create monitoring dashboards
• Establish escalation procedures

Interview Tips

• Explain the six dimensions
• How do you implement quality checks?
• Describe a data quality incident