Saniamo Mercy
Monitoring, Evaluation, Research and Learning
I design measurement and learning systems that help programmes use evidence to improve implementation.
Monitoring, Evaluation, Research and Learning
I design measurement and learning systems that help programmes use evidence to improve implementation.
I design measurement and learning systems for programmes, and use data and evaluation to understand what is changing, what is not, and what teams may need to do differently.
I am interested in how evidence gets used during implementation, not only what gets reported at the end. That means asking whether the data is good enough to support a decision, whether the assumptions behind a programme still hold, and what the evidence suggests should change.
Good measurement should make clear what the evidence can support, what it cannot, and where the uncertainty matters for a decision.
The programme wanted to concentrate additional support on the businesses most likely to create jobs, but there was no agreed way to identify them or to test whether the extra support actually worked.
Led the M&E design: theory of change, measurement approach, indicator framework and baseline design.
Worked backwards from the intended outcomes to map the causal pathways, made the assumptions explicit, and connected each pathway to indicators and data sources. Designed the M&E plan, logframe and indicator performance tracking table, and built the application and baseline surveys used to collect standardised data, including consent design compliant with Kenya's Data Protection Act.
A programme hypothesis that could be tested, with a measurement system designed to distinguish between what the data could establish and what it could not.
The system let the programme test its assumptions about who benefits from additional support, rather than waiting until the end to find out. It also surfaced response-rate limitations and duplicate records beneath the headline results, which changed how the programme reported the findings and collected data in the following cycle.
A Rwandan development bank's SME finance programme was producing plenty of data, but stakeholders could not see progress against targets without manual work each reporting round.
Rebuilt the M&E reporting dashboard used by World Bank and BRD stakeholders.
Restructured how SME performance indicators were defined, collected and consolidated, then rebuilt the reporting layer so the underlying data and the reported figures stayed in step.
A reliable, decision-ready view of programme progress that stakeholders could interrogate themselves rather than request.
Teams could see progress against targets during implementation, spot issues earlier and ask their own questions of the data instead of waiting for the next reporting cycle.
The programme needed evidence of business growth that would hold up to donor scrutiny, from data with the uneven quality typical of self-reported business figures.
Designed and ran the analysis behind the annual impact report for CapPlus and the Mastercard Foundation.
Built a matched-pair analysis using compound annual growth rate methodology with interquartile-range outlier removal, so that a handful of extreme values could not carry the headline result.
Growth findings the programme could defend, alongside a clear account of what the data could and could not establish.
A more defensible basis for interpreting business growth, with the limitations of the evidence made explicit alongside the findings.
A financial inclusion programme had activities and indicators but no explicit account of how the intervention was expected to produce change, or what had to be true for it to work.
Built the theory of change as a standalone deliverable for the client, BFA Global.
Mapped the causal pathways and surfaced the assumptions sitting underneath them, drawing on earlier randomised-trial impact analysis of previous cohorts rather than starting from scratch.
A clear account of how the programme expected change to happen, the assumptions that needed to hold, and what was worth measuring during implementation.
I write about measurement, adaptive management, evaluation, and how evidence gets used in practice.
On how tracking systems reshape the behaviour they were meant to observe, and what happens when external measures start to replace internal judgment.
Strategy is a hypothesis, not a contract with the future. Every strategy rests on assumptions, and assumptions have a shelf life.
More at @everythingdata on Medium.