Most MEL frameworks in the development sector are designed to satisfy donor reporting requirements, not to drive programme improvement. Here is how to build systems that do both.
Monitoring, Evaluation and Learning (MEL) occupies an uncomfortable position in development programmes. It is universally required, frequently under-resourced, and rarely designed with learning as its primary purpose. The result is a vast quantity of data that satisfies donor compliance requirements and generates very little actionable insight for programme teams.
This is a structural problem, not a technical one. Fixing it requires rethinking what MEL systems are for before designing the tools and processes that implement them.
The compliance trap
Most donor-funded MEL frameworks are built backwards: from the indicators in the log frame, through the data collection instruments needed to measure those indicators, to the reporting schedule that determines when data must be available. This produces systems that are very good at generating periodic reports and very poor at answering the questions programme teams actually face in the field.
The questions that matter — Is this intervention reaching the most marginalised households? Is the training changing practice, or just knowledge? Are there unintended negative consequences we are not tracking? — rarely appear in a standard log frame. They require adaptive, real-time data collection, not quarterly indicator updates.
Building systems that drive learning
A MEL system designed for learning starts with a different question: what do programme managers need to know, and when do they need to know it, to make better decisions?
This shifts the design logic from reporting schedules to decision cycles. It asks what decisions will be made at six months, at eighteen months, at programme close — and works backwards to determine what evidence needs to be available at each point.
Practically, this means:
- Separating the learning system from the accountability system. Routine programme data goes to the learning system in near-real-time. Audited impact data goes to the accountability system on the donor reporting schedule. Both are necessary; they should not be the same system.
- Investing in sense-making sessions, not just data collection. Data without interpretation is noise. Quarterly reflection sessions where programme staff, community representatives, and MEL officers review emerging data together are more valuable than many organisations realise.
- Tracking what you will actually use. A sixty-indicator framework is a red flag. If programme management cannot use sixty indicators to make decisions, they will not — and data collection becomes a compliance exercise that drains staff time without improving anything.
The role of technology
Digital data collection tools (ODK, KoboToolbox, CommCare) have dramatically reduced the cost of field data collection. They have not solved the learning problem, because the problem is not data scarcity — it is data utilisation. Organisations that invest heavily in data collection tools without investing equally in analytical capacity and sense-making processes tend to end up with better data that is used less.
The right technology investment supports a learning culture, not just a reporting function. That means dashboards that programme managers actually look at, alert systems that flag emerging issues before the next quarterly report, and visualisation tools that make community-level data legible to community members — not just to analysts in head office.
Topics