Evaluation design, data collection, third-party monitoring and performance audits for development and humanitarian programmes.
Funders no longer accept a narrative report as evidence. What they ask for is a defensible account of what changed, how you know, and what you would do differently — and the gap between an organisation that can produce that and one that cannot is rarely capability. It is usually that nobody set up the measurement before the programme started, and by the time the report is due the window has closed.
Most of the evaluations we are asked to rescue have the same shape. Activity counts are excellent: sessions delivered, people reached, kits distributed. Outcomes are asserted rather than measured, because the indicators were written to sound good in a proposal rather than to be collected by a field officer with a phone and forty minutes. Nobody notices until an evaluator asks where a number came from.
We work at both ends. Early, that means designing indicators someone can actually collect, taking a baseline while it is still possible, and building a data trail that survives staff turnover. Late, it means independent evaluation against the OECD-DAC criteria, third-party monitoring where a funder needs verification they did not commission from the implementer, and performance audits that say plainly which findings are solid and which are indicative.