Evidence
Evidence,
on the record
What we publish, when we publish it, and what we are prepared to say we do not know. A programme record is only worth reading if the standard behind it is visible.
Our standard
A number without
a method is
an opinion
Impact reporting has a credibility problem, and it is largely self-inflicted. Reach figures count attendance. Round numbers appear without a baseline. Changes are attributed to a programme with no comparison group and no acknowledgement that something else might have caused them.
None of that survives contact with an auditor, and increasingly none of it survives contact with a disclosure requirement either. So we hold to four rules, and we would rather publish a smaller number that holds than a larger one that does not.
Four rules
No outcome without a baseline
If we did not measure the starting position, we do not report a change. Where a baseline was not possible, the reporting states that and describes what we observed instead.
Attrition is a published number
Participants lost between intake and endline are reported as a count and a proportion. Results are not quietly recalculated on whoever remained.
Attribution is claimed only where it is earned
Where there is no control or comparison group, we report the measured change and say explicitly that we cannot separate our contribution from everything else happening in that district.
Underperformance is published alongside results
Every programme review states what did not work and what we would change. A record with no failures in it has not been examined.
Programme record
Live status for every programme. Figures are added here as each endline completes and its verification pack is assembled — we do not publish interim numbers, and we do not publish reach figures in place of outcomes.
| Programme | Area | Primary indicator | Baseline | Endline | Status |
|---|---|---|---|---|---|
| Maahwari | Menstrual health & circular products | Sustained access at follow-up (SDG 6.2) | Complete | In field | Reporting on completion |
| Maahwari | Education retention | School days retained per cycle (SDG 4.1) | Complete | In field | Reporting on completion |
| Climate resilience | Flood response — hygiene & water | Households reached within 30 days (SDG 13.1) | Complete | In field | Reporting on completion |
| Waste displacement | Circular economy | Single-use waste displaced per participant (SDG 12.5) | Method published | Pending reuse data | Method only |
Where a row shows “Method only”, it means we have published how a figure would be calculated but have not yet collected the data that would let us state it. We would rather show the gap than fill it with an estimate.
What a verification
pack contains
Assembled at close for every programme, and complete enough that a sceptical reader could disagree with our conclusion using our own data.
- The problem statement agreed at scoping, including stated measurement limits
- Full indicator set with SDG target and national framework mapping
- Baseline: population, instrument, sample, date and method
- Endline: the same, plus interval and comparison basis
- Raw instruments as administered, in the language used
- Every exclusion, with the reason for it
- Every adjustment, with the reasoning behind it
- Cost per outcome modelled at design, restated against actuals
- Programme review, including what underperformed