Home/Guides/What is a MEL system

Reference guide

What a MEL system is and how it differs from traditional M&E

What is a MEL system?

A MEL system is the infrastructure an organization uses to monitor its program, evaluate it and learn from the results in order to adjust it. It integrates six components: Theory of Change, indicators, collection instruments, data sources, data pipeline and reporting. The difference from traditional M&E is that learning is an explicit component, with defined routines and owners.

MEL stands for monitoring, evaluation and learning. The three functions answer different questions. Monitoring answers what is happening now. Evaluation answers whether it worked and why. Learning answers what we are going to do differently as a result.

The key word is system. An organization can have indicators, carry out evaluations and produce reports without having a MEL system: if each piece is built separately and has to be consolidated by hand every cycle, what exists is a set of deliverables, not an infrastructure.

The six components

1. Theory of Change

The program's causal logic: which resources go in, which activities are carried out, which outputs they generate, which outcomes they produce and which impact they contribute to. It makes explicit the assumptions that hold each link, which is what makes it possible to know where the chain failed when the result does not appear.

2. Indicators

The observable variables that show whether each outcome is occurring. An indicator without a definition, formula, source, frequency and owner is not an indicator: it is an intention written into a logical framework.

3. Collection instruments

The questionnaires, forms, interview protocols and administrative records that produce the data. They are designed from the indicators, not the other way around, and are piloted before going into the field.

4. Data sources

Where each piece of data originates before entering the pipeline: own forms, management platforms, administrative records, third-party surveys. Mapping them is what makes it possible to know, for each indicator, which source feeds it and whether that source is reliable.

5. Data pipeline

The path the data travels from the moment it is collected until it reaches the report: source integration, cleaning, validation and storage. It is the component that eliminates the most manual work and the one almost no organization has solved.

6. Reporting

A single source of truth that feeds the funder, the board and operations with the same figure. When each audience receives a version built separately, inconsistencies appear in the first audit.

The learning routine is the one almost no one builds. How often the evidence is reviewed, who takes part in that review and where the decision that was made is recorded. Without it, correct findings end up filed away.

MEL, M&E and reporting dashboard: how they differ

The three terms are used as synonyms and they are not. The practical difference lies in what is left working when the project closes.

Criterion MEL system Traditional M&E Reporting dashboard
Purpose Decide, learn and be accountable Be accountable to the funder Visualize figures that are already consolidated
Who uses it Implementation team, management and funder MEL specialist and funder Management, in the monthly meeting
Frequency of use Continuous, integrated into operations Per reporting cycle Variable, often declines after launch
What it produces Documented decisions and adjustments to the program Reports and progress reports Charts
What happens with the findings They enter a routine with an owner and a deadline They stay in the final report There are no findings, there are figures
What is left at close Documented infrastructure and a team able to run it An archive of reports A dashboard that goes out of date

The dashboard is not a bad product: it is a component of the system presented as if it were the whole system. A dashboard connected to data that someone still consolidates by hand did not solve the problem, it only moved the work somewhere else.

When an organization needs a MEL system

Not all of them do. These are the concrete signs that it does:

  • The reporting cycle consumes between 40 and 100 hours of manual work and is repeated in full every quarter.
  • There are more than three data sources that do not talk to each other.
  • Indicators committed in the logical framework reach the report with incomplete data.
  • Different audiences receive different figures about the same thing.
  • An audit found inconsistencies between versions of the same data point.
  • The organization funds several implementers and each one measures with its own criteria.
  • The MEL team turned over and the system's memory left with the people.

If none applies, what is probably missing is a specific component, not a complete system. Starting with the right component costs less and works sooner.

The standards that frame it

A rigorous MEL system rests on established frameworks, not on criteria invented by the provider:

  • The OECD DAC evaluation criteria: relevance, coherence, effectiveness, efficiency, impact and sustainability. They are the reference framework of international cooperation.
  • The UNEG norms and standards, the evaluation group of the United Nations system, for independence, credibility and utility of evaluation.
  • The GIIN IRIS+ catalog, for impact metrics comparable across organizations and investors.

Using them has a practical advantage beyond rigor: when the indicator is defined according to a public catalog, the discussion with the funder about whether the figure is valid gets considerably shorter.

Frequently asked questions

Next steps. If the component you are missing is the causal framework, continue with the Theory of Change guide. If the problem is the indicators, go to MEL indicators. If you want to diagnose the complete system, the measurement kit has the five-dimension checklist.

Does your system have all six components?

Write to us with your specific situation and we will reply within 48 business hours with our read of the problem.