Data quality and review states

Four states

• VALID — the record is complete and consistent
• WARNING — usable, but something should be checked
• REQUIRES REVIEW — geometry or a critical value could not be read safely
• INVALID — a mandatory value is missing or contradicts the rules

What raises a warning

Each issue is recorded against the field that caused it, with a code and a plain-language message — an unrecognised class, a projected coordinate supplied without a CRS, geometry that could not be parsed, or an asset type not previously used with that class.

Why it matters

Data quality is the difference between a register that supports decisions and one that merely exists. The data-quality dashboard shows where the gaps are, and each exception links back to the record that needs correcting.