MEL systems that survive the second year

Monitoring and evaluation platforms have a survival problem. Year one looks great — clean dashboards, board-ready charts, donor approval. Year two, the field team stops entering data, the dashboard quietly goes stale, and the programme reverts to Excel. We have seen this enough times to name what separates the systems that survive from the ones that don't.

The four design choices that matter 1. **Data entry is the field team's tool, not the donor's tool.** If the form is designed for the report and not for the worker, the worker stops filling it. 2. **Offline-first is non-negotiable.** Connectivity is the enemy. A system that punishes the field team for working in a poor-signal area will lose to a paper form. 3. **Reporting is auto-rolled, not hand-rolled.** If a programme manager has to copy numbers from the dashboard into a Word document every month, the system has failed at its actual job. 4. **A single identity spine across programmes.** A foundation that runs three programmes should not run three MIS. Shared identity, shared geography, shared reporting — modular programme logic on top.

The "year two test" The honest test of a MEL platform is not the launch demo. It is whether the field team is still entering data in month fourteen. If they are, the design choices above are usually present. If they are not, at least one of them is missing.

The bigger framing MEL is not a reporting product. It is an operational system for a programme. The systems that survive treat it that way from day one.