MEAL Demystified: What Every NGO Should Know

If you work in the development sector, you will have encountered the acronym MEAL. It appears in job titles, donor requirements, proposal templates, and countless training programmes. Understanding how MEAL evolved over the decades helps explain why it looks the way it does today. But ask ten practitioners what MEAL actually means in practice, and you'll likely get ten different answers.

That confusion is no accident. MEAL is both simpler than people make it sound and subtler than most introductions suggest. This guide aims to cut through the jargon and explain what each component actually involves, and why all four matter.

Breaking down the acronym

MEAL stands for Monitoring, Evaluation, Accountability, and Learning. Each component serves a distinct purpose, though in practice they overlap and reinforce each other. The evaluation half of M&E draws on a shared vocabulary, and the best known is the OECD Development Assistance Committee's evaluation criteria (relevance, coherence, effectiveness, efficiency, impact and sustainability), revised by the OECD in 2019 and widely used in development co-operation, and most major donors now expect evaluations to engage with them.

[Illustration 1: The MEAL cycle]

Monitoring: the foundation

Monitoring is the continuous, routine collection of data about your programme's activities and outputs. Think of it as keeping your finger on the pulse of implementation.

Good monitoring answers questions like: How many training sessions have we conducted? How many participants attended? What percentage were women? Are we on track to meet our targets? Where are we falling behind? Of course, the value of monitoring depends on data quality, a challenge that looks very different in the field than in the classroom.

The key distinction is that monitoring focuses on what you're doing (activities and outputs), while evaluation asks what difference it makes (outcomes and impact). You might monitor that you've distributed 1,000 bed nets, but monitoring alone won't tell you whether malaria rates have declined. Establishing that link takes evaluation. The Cochrane systematic review of insecticide-treated nets (Pryce, Richardson & Lengeler, 2018), which included 23 randomised trials, found that nets cut all-cause child mortality by 17% compared with no nets (rate ratio 0.83, from five of those trials covering 200,833 participants). No distribution log could have shown that.

Evaluation: the deeper look

While monitoring tracks implementation, evaluation assesses results. It asks whether your programme is actually creating the changes you intended, and often, whether those changes would have happened anyway.

Evaluations can be formative (conducted during implementation to improve the programme) or summative (conducted at the end to assess overall results), a distinction the philosopher Michael Scriven introduced in his 1967 essay "The Methodology of Evaluation" and that has anchored the field for over fifty years. They can focus on process (how well the programme was implemented), outcome (what changes occurred), or impact (what changes can be attributed to the programme).

Impact evaluation, in particular, requires thinking carefully about the counterfactual: what would have happened without your intervention? As J-PAL puts it, you cannot go back in time to see what the same households would have experienced without the programme. Evaluators therefore build a comparison group, and J-PAL's own method is random assignment. Approaches like quasi-experimental designs offer practical ways to approximate counterfactuals in field settings. This is where methodologies like randomised controlled trials, difference-in-differences, or regression discontinuity become relevant.

Accountability: the two-way street

Accountability is perhaps the most misunderstood component of MEAL. Reporting to donors matters, but accountability starts with answerability to the communities you serve.

"Accountability means the people we're trying to help have real power to influence our work, beyond receiving it."

This framing is now codified in sector standards, for example in the Core Humanitarian Standard (2024 edition), whose glossary defines accountability as "the process of using power responsibly, and taking account of and being held accountable by different stakeholders, primarily those who are affected by the exercise of such power." It adds that accountability means putting people and communities at the centre of decisions on issues that affect them. Meaningful accountability includes mechanisms for communities to provide feedback, lodge complaints, and participate in decisions that affect them. Effective community feedback loops are at the heart of this. It requires transparency about what you're doing, why you're doing it, and what resources you're working with.

Too many organisations treat accountability as a compliance checkbox when it is a commitment to power-sharing. The result is feedback mechanisms that exist on paper and never influence programming.

[Illustration 2: The accountability relationship]

Learning: closing the loop

Learning is what transforms data into improvement. It's the "so what?" that connects all the monitoring data and evaluation findings to actual changes in how you work.

Organisations often collect extensive data and conduct elaborate evaluations, only to file the reports away and continue doing exactly what they were doing before. Filing the report is documentation, and learning begins only when the next decision changes. As 3ie has argued, for development institutions, learning requires more than collecting data. At the 3ie panel, Håvard Mokleiv Nygård of the Norwegian Agency for Development Cooperation said: "We collect lots of results data…what we know, though, is that there's something wrong."

Learning needs dedicated time for reflection, psychological safety to admit what isn't working, and routines that feed insights into decisions. Failure becomes information when nobody is punished for reporting it.

Why all four matter

A programme can fail at any one of the four. Some organisations do excellent monitoring but never step back to evaluate whether their activities create real change. Others conduct rigorous evaluations but don't have the routine data to understand implementation. Many tick the accountability boxes while communities still have no real say. And most struggle to convert any of it into systematic learning.

MEAL only works as an integrated system. Monitoring feeds evaluation by providing the routine data that evaluations build upon. Evaluation informs accountability by giving communities and other stakeholders substantive information about results. Accountability enriches learning by surfacing perspectives that internal analysis might miss. And learning closes the loop by improving all three.

Getting started

If your organisation is just beginning to build MEAL capacity, start with the basics: What questions do you actually need to answer? What decisions will you make differently with better evidence? Who are you accountable to, and what do they need to know?

Don't try to build a perfect system from scratch. Start simple, learn what works in your context, and iterate. A MEAL system is good when it helps you do better work, whatever it looks like on paper.

ImpactMojo's MEAL courses and studios (the free MEL Basics 101 deck, the flagship The Evidence Question course, and the MEL Studio) are designed to help you build these skills step by step, with examples drawn from the South Asian development context. Every practitioner deserves access to the fundamentals.