Every development project collects data.

It collects attendance records, household surveys, service-delivery figures, training assessments, feedback forms, budget trackers and outcome indicators. It counts participants, maps locations, measures outputs and, increasingly, visualises results through dashboards.

Yet one question is asked far less often than it should be: what changed because of this data?

This is not a rhetorical question. It goes to the heart of how development programmes are managed.

Across the sector, organisations invest substantial time and resources in data collection. Field staff administer surveys. Enumerators travel long distances. Community members answer repeated questionnaires. MEAL teams clean datasets, reconcile inconsistencies, calculate indicators and prepare reports. By the time a dataset reaches a decision-maker, it has already required money, labour, trust and attention.

And too often, it is used to describe the past rather than shape the future. That is not merely an efficiency problem. It is an accountability problem.

The hidden cost of unused evidence

Data is often treated as if it were free once it has been collected. It is not.

Every indicator carries a cost. Someone had to understand the question, record an answer, transmit information, verify it and enter it into a system. In rural or hard-to-reach areas, this effort may involve unreliable connectivity, difficult transport, limited electricity and staff who are already carrying multiple responsibilities.

For communities, the cost is even more personal. People give their time. They share information about their income, health, livelihoods, household conditions or access to services. They do so with an implicit expectation: that their experience will matter.

When the information disappears into a spreadsheet, a quarterly report or a donor presentation without influencing a decision, that expectation is broken.

The issue is not that reporting is unnecessary. Reporting matters. Donors, governments, implementing partners and communities all have a legitimate interest in knowing what has been achieved. The problem begins when reporting becomes the final destination.

A strong MEAL system should not end with the submission of a report. It should begin a conversation.

Madagascar’s challenge is not a lack of information

Madagascar is not starting from zero. National monitoring and evaluation mechanisms exist, and recent assessments have identified both the foundations already in place and the areas that still require investment.

The MESA diagnostic on Madagascar’s national monitoring and evaluation capacities, reflected in a UNICEF Madagascar summary, points to familiar challenges: incomplete data, delays in reporting, uneven access to results, fragmented information systems and limited use of evidence in some decision-making processes.

These are not uniquely Malagasy problems. They appear across the development sector, in countries with far more advanced digital infrastructure and far larger data teams. But Madagascar makes the challenge especially visible.

The country combines significant development needs with geographical diversity, uneven connectivity, difficult access to some areas and a wide range of actors collecting information for different purposes. Government institutions, NGOs, donors, social enterprises, humanitarian programmes and community organisations may all be gathering data at the same time, often with different indicators, tools and reporting calendars.

The result can be a paradox: a project may be surrounded by data while still lacking the information needed to make a timely decision.

This is why the most important question is not, “Do we have enough data?” It is, “Do we have the right information in time to act?”

A dashboard cannot explain “why”

Digital tools are changing the possibilities for monitoring, evaluation and learning.

KoboToolbox can reduce paper-based collection. Excel can structure operational tracking. DHIS2 can strengthen health information systems. Power BI and other business-intelligence tools can make patterns visible more quickly than traditional narrative reporting.

These tools matter. But they do not replace judgement.

A dashboard can show that participation has declined in a particular district. It can compare municipalities, flag missing data or reveal a sudden change in a trend. It cannot reliably explain why.

Why are adolescent girls leaving a vocational training programme? Why are community health workers submitting fewer reports? Why did an agricultural practice succeed in one commune but fail in another? Why are people attending awareness sessions but not changing behaviour?

The answers are rarely contained in a graph. They emerge from field observation, conversations with communities, operational knowledge and the willingness of programme teams to listen to evidence that may challenge their assumptions.

This is where many MEAL systems become weaker than they appear. They are technically capable of collecting and displaying data, but they are not organised to interpret it collectively. The real value of a dashboard is not the dashboard itself. It is the conversation it enables.

From compliance to learning

A meaningful MEAL system operates as a cycle, not a production line.

It gathers evidence. It brings people together to interpret what that evidence means. It supports a decision. It observes whether the decision produced the intended result. Then it learns and adjusts again.

This may sound obvious. In practice, it requires discipline: the same discipline the ex-USAID CLA framework (Collaborating, Learning and Adapting) asks of implementing partners: learning continuously, with partners, in order to adjust course while a project is still running, not only once it ends.

It requires managers to create time for teams to look at evidence before key decisions are made, not after. It requires field staff to be treated as interpreters of data, not simply collectors of it. It requires community feedback to be considered alongside quantitative indicators. And it requires leaders to make one of the hardest moves in organisational life: changing direction when the evidence suggests that the original plan is no longer working.

The United Nations Development Programme’s work with 14 African country offices offers practical examples of this shift. The initiative highlights feedback loops, community engagement, digital systems, storytelling and adaptive learning as interconnected parts of stronger impact measurement. It also describes how a digital monitoring and evaluation system first developed in Burundi was replicated in Madagascar to support data quality, visualisation and faster decision-making.

The lesson is not that every organisation needs a more advanced platform. The lesson is that data systems should be designed around decisions, not around reports.

The digital question is also an inclusion question

It is tempting to see digital transformation as the obvious answer. But digital systems only improve decision-making when they fit the realities of the people expected to use them.

At the end of 2025, Madagascar had an estimated 6.71 million internet users, equivalent to 20.4 percent of the population, according to DataReportal’s Digital 2026: Madagascar report. Around 79.6 percent of the population remained offline.

These figures do not argue against digital tools. They argue against designing systems as if connectivity, devices, electricity, digital literacy and access to dashboards were equally available everywhere.

A monitoring system designed from a capital city can look efficient on paper while creating additional burdens for a field team working with intermittent network coverage. A mobile form may simplify collection in one area and slow it down in another. A sophisticated dashboard may be useful for senior management but inaccessible to the communities whose experiences generated the data.

The question, therefore, is not whether a system is digital. It is whether it is useful, inclusive and appropriate.

Sometimes the best learning process is a Power BI dashboard reviewed during a management meeting. Sometimes it is a printed map discussed with community representatives. Sometimes it is a short phone call with a field coordinator who has noticed something no indicator captured. Good MEAL practice recognises the value of all three.

Skills are not enough; operational capacity matters

The development sector has no shortage of training. There are workshops on logframes, indicator design, data quality, survey tools, visualisation and evaluation methods.

But training alone does not guarantee institutional capacity.

A participant may complete a course on Power BI and still struggle to identify which indicators matter. A team may know how to build a KoboToolbox form and still collect information that no one uses. An organisation may have a well-designed results framework but lack a routine for discussing findings and adapting implementation.

The missing link is operational capacity: the ability to connect management questions with data needs, collection systems, data quality, analysis, visualisation and decision-making.

This is the space in which locally rooted initiatives such as Helpazone can make a practical contribution. By linking MEAL, data analytics, business intelligence and programme management with real organisational needs, Helpazone focuses on helping NGOs, social enterprises and operational teams move beyond tool familiarity towards usable systems, practical outputs and continued support after training.

The goal should not be to produce more certificates. It should be to ensure that, when a programme manager asks, “What should we do differently next month?”, the team has the evidence, confidence and systems needed to answer.

The real test of a MEAL system

The quality of a MEAL system should not be measured by the number of indicators it tracks. Nor should it be measured by the size of its database, the sophistication of its dashboard or the length of its report.

A better test is simpler:

  • Did the data reach the people who needed it?
  • Did they understand what it meant?
  • Did it lead to a decision?
  • Did the organisation learn from the result of that decision?
  • Did communities see that the information they shared had value?

If the answer is yes, then data becomes more than an administrative requirement. It becomes part of how an organisation thinks.

Madagascar’s experience offers a wider lesson for the development sector. The future of MEAL is not simply more data, faster collection or better visualisation. It is the ability to turn evidence into action while there is still time to make a difference.

The most expensive data is not the data that takes the longest to collect. It is the data that no one uses.

Sources