IATI Data Quality Framework — consultation

This repository collects feedback on the draft IATI Data Quality Framework, which is being used to assess the quality of the data organisations publish to IATI.

Comments are open until 17 November 2026.

How to comment

Each part of the framework has its own issue, so that comments stay attached to the thing they are about:

• one issue for each of the 11 components of the four dimensions — availability, timeliness, coverage and comprehensiveness
• one issue for each of the 33 indicators that make up those components

Browse the issues and comment on the ones you have a view on. You do not need to comment on all of them, and you are welcome to comment on only your own area of interest.

Each issue describes what is being measured, the specific tests behind it, and any caveats that are still unresolved. Useful things to tell us:

• whether what is being measured is reasonable to expect of publishers
• whether the tests are a fair way of measuring it
• whether the thresholds for Good, Very good and Excellent are set at the right level
• whether an indicator should apply to every organisation, or whether there are organisation types it should not apply to

Comments are public. You will need a free GitHub account.

Before you comment

The dashboard explains how an assessment is reached, lists every indicator and its thresholds, and flags what is still provisional:

• How the assessment works
• Reimagining Data Quality: A User Centric Approach — Proposal for Consultation

Organisations publishing to IATI can also sign in to the dashboard to see their own preliminary assessment.

If you cannot use GitHub

If you would rather not use GitHub, want to share feedback privately, or have a question rather than a comment, email support@iatistandard.org.

What happens next

Feedback will be reviewed and used to finalise the framework — the indicators, the thresholds and the weightings — before the dashboard is published. A snapshot of the final assessments will feed into IATI's State of the Data report, and publishers will be consulted on that before it is published.

For maintainers

The issues in this repository are generated rather than written by hand, so that they say the same thing as the dashboard. The generator lives in the dashboard frontend repository, at scripts/consultation/:

python3 scripts/consultation/generate_issues.py   # build the issue texts
python3 scripts/consultation/post_issues.py       # dry run — shows what would be posted
python3 scripts/consultation/post_issues.py --post

It skips any issue whose title already exists, so it is safe to re-run. It does not update issues that have already been posted: if wording changes after an issue exists, edit that issue here.