Illustrative: "Indicator 'Women in farmer-group leadership positions': no result reported for 2026-Q3" — the exact, specific gap, not a generic warning.
Governance & Performance
What data quality means, and why it should be a routine check, not a crisis review
Most damaging data problems are not fraud — they are ordinary gaps that go unnoticed until a report is due and it is too late to fix them cheaply.
Step 1 · Learn
Understand the concept
Data quality in programme monitoring usually reduces to a small set of concrete, checkable failure modes: values missing entirely for a period, values that are implausible on their face (a percentage above 100%, a negative count), inconsistent definitions between reporting periods, and gaps in coverage (a site or partner that never reported at all rather than reporting a genuine zero). Catching these routinely and early is dramatically cheaper than discovering them during report finalization or during an external review.
A genuinely useful data-quality check is deterministic and specific — "indicator X has no target set for period Y" — rather than a vague, unexplained warning. A specific flag tells the person responsible exactly what to fix; a generic one just adds anxiety without direction.
How METRA GET supports this
Data Quality flags in METRA GET are deterministic checks against real gaps: an indicator with no target set for the period, one with no result reported, an incomplete semi-annual aggregation, or a project with zero budget lines or zero reach on record.
A report only ever points to its own Data Quality section if that section is actually included in the report — it never tells a reader to "see the section below" when that section was left out.
From concept to your own project
Data Quality rules run automatically on real project data the moment it's entered — available immediately, with nothing to configure.
Sign in to work with your organization’s real data, or request a demo to see it walked through.
Governance & Performance