Monitoring and evaluation (M&E) is the practice of tracking what a program does while it runs, and judging whether it worked once there is enough evidence to judge. Monitoring answers whether delivery is on track. Evaluation answers whether the program made a difference, and why. An M&E system is what makes both possible.
What is monitoring and evaluation?
M&E is how a program knows what it is doing and whether it is working. The two halves answer different questions on different schedules.
Monitoring is continuous. It tracks activities, outputs and progress against plan, usually monthly or quarterly, and its purpose is management: catching problems early enough to act on them.
Evaluation is periodic and deeper. It asks whether the program achieved what it set out to achieve, whether the change can reasonably be attributed to the program, and what should be done differently. It usually happens at midpoint and at the end, sometimes years later.
Neither substitutes for the other. Monitoring without evaluation produces a program that reports activity forever without knowing whether any of it mattered. Evaluation without monitoring produces an evaluator with no data to work from. For the difference in detail, see monitoring vs evaluation.
What does M&E stand for, and what about MEL and MEAL?
M&E stands for monitoring and evaluation. Two longer acronyms are in common use and mean slightly different things.
| Acronym | Stands for | What the extra letter adds |
|---|---|---|
| M&E | Monitoring and Evaluation | The base pair |
| MEL | Monitoring, Evaluation and Learning | Makes explicit that findings must change decisions, not just be reported |
| MEAL | Monitoring, Evaluation, Accountability and Learning | Adds obligations to the people the program serves, including feedback and complaint mechanisms |
The choice is usually organizational convention rather than a real difference in method. What matters is whether the learning and accountability functions actually exist, not which letters appear in the team's name. See MEL vs M&E vs MEAL terminology for how the three are used in practice.
What does an M&E system contain?
An M&E system is the set of things that have to exist before monitoring or evaluation can happen. Most systems contain the same components, whatever they are called locally.
| Component | What it is | Where to read more |
|---|---|---|
| Results logic | The chain from activities to outputs to outcomes to impact, stating what the program expects to change | Results chain |
| Framework | The document specifying what will be measured, how, by whom and how often | M&E framework |
| Indicators | The specific measurable variables that track progress against each result | Indicator |
| Data collection | The instruments, sampling approach and field process that produce the numbers | Survey design |
| Data management | Where data lives, who can change it, and how quality is checked | Data quality assurance |
| Reporting and use | The reports, reviews and decision points where findings actually change something | MEL plans |
| Evaluation plan | What will be evaluated, when, by whom, and against which questions | Evaluation TOR |
Programs commonly have the first three and are thin on the last three. A framework with good indicators and no decision point where findings get used is a reporting system, not an M&E system.
What M&E tools are used?
"M&E tools" covers three different things, and the ambiguity causes real confusion when people compare notes.
Data collection tools are the instruments themselves: questionnaires, interview guides, observation checklists, and the mobile platforms used to administer them in the field.
Analysis and tracking tools are what hold the data once collected: indicator trackers, dashboards, statistical software and the spreadsheets most programs actually run on.
Planning tools are the documents that structure the system: the logframe, the results framework, the indicator reference sheet, the MEL plan.
A program asking which M&E tool to buy usually needs the second kind. A program asking which M&E tool to use usually needs the first or third.
Who does M&E on a program?
Responsibility is normally split, and the split matters more than the job titles.
Program staff collect most of the data, because they are the ones in contact with activities and participants. Their reporting burden is the single most common reason M&E systems fail in practice.
An M&E officer or team designs the system, checks quality, aggregates and reports. On small programs this is a part-time share of one person's role rather than a post.
External evaluators conduct formal evaluations. They are commissioned deliberately, on written terms of reference, and independence is the reason to use them.
Program leadership decides what happens as a result. This is the role most often left undefined, and it is the one that determines whether the system is worth running.
How do you set up M&E for a program?
In rough order, and each step depends on the one before it.
- State the results logic first. What is expected to change, for whom, and through what mechanism. Everything downstream is derived from this, and skipping it produces indicators nobody can interpret.
- Choose indicators against the logic, not against a list. Each one should tell you whether a specific result is happening. Fewer, better-defined indicators beat comprehensive coverage.
- Define each indicator properly before any data is collected: the exact wording, unit, disaggregation, source and collection frequency. This is what an indicator reference sheet is for.
- Design collection around the analysis you intend to do, working backward from what you will report rather than forward from what seems worth asking.
- Decide where findings get used. Name the meeting, the report or the review where each piece of information changes a decision. Anything that changes no decision should be cut.
- Plan the evaluation separately. It has different questions, a different timeline and usually a different author than the monitoring system.
The Evaluation Readiness tool checks whether a program is ready for step 6, and the Method Selector helps with step 4.
How M&E differs from related work
The term gets used loosely, and three neighbours are commonly confused with it.
| What it is | How it differs from M&E | |
|---|---|---|
| Research | Generates knowledge that generalizes beyond one program | M&E answers questions about this program, for this program's decisions. Research asks what is true in general |
| Audit | Verifies that money and process complied with rules | M&E asks whether the work achieved anything. A program can pass an audit and have accomplished nothing |
| Performance management | Manages the performance of staff and the organization | M&E measures the program. The two use similar language and answer to different people |
The practical test is who the answer is for. If a finding cannot change a decision someone in the program is going to make, it is probably not M&E.
What M&E looks like in practice
A three-year livelihoods program. Monitoring runs monthly: households reached, training sessions delivered, inputs distributed, tracked against plan by field staff and aggregated by an M&E officer. A baseline survey at start and an endline at close measure income and food security against targets. A midterm review, internal, asks whether delivery is on track and what to change. A final external evaluation asks whether incomes actually rose, whether the program can claim credit, and what should be done differently next time.
A one-year advocacy project. Monitoring is lighter: meetings held, submissions made, coalition members recruited. There is no meaningful baseline survey, because the outcome is a policy change rather than a measurable population characteristic. Evaluation is therefore qualitative and contribution-focused, asking what changed in the policy environment and how plausibly the project contributed. Forcing this program into an experimental design would produce a worse answer, not a better one.
The difference between the two is not budget or ambition. It is whether the outcome is countable and whether a counterfactual is available.
Common mistakes
Measuring everything. Indicator lists grow because adding one is easier than defending its removal. The cost lands on field staff and the data quality falls across the board.
Confusing outputs with outcomes. Counting trainings delivered is monitoring. Knowing whether anyone changed what they do is evaluation. Programs that report only the first often believe they have evidence of the second.
Building the system after the program starts. Baseline data cannot be collected retrospectively, and without it most evaluation questions become unanswerable.
Treating M&E as a reporting obligation. A system designed to satisfy a donor report will satisfy the donor report and inform nothing. The test is whether any decision would have been made differently without it.
Related topics
The monitoring vs evaluation entry covers the distinction in detail. M&E framework and M&E system design cover the document and the infrastructure respectively. For building the plan itself, start at MEL plans.