Community needs assessment: finding out what really matters before the budget is allocated

1. The problem: a small budget and a long list of problems

A district town hall has money for one large project next year. Councillors want the main road repaired, the head teacher asks for the sports hall to be refurbished, the family doctor talks about older people living alone who never make it to appointments, and a local NGO argues for a youth centre. Every proposal is legitimate, every one has supporters, and nobody has comparable evidence. A needs assessment is the research that answers the question "which problem affects the most people, most seriously, and where would an intervention make the biggest difference?".

NGOs preparing a project face the same question, as do donor-funded programmes that must justify their choice of communities and activities, and public institutions planning social services. The scale differs; the logic does not.

2. What a needs assessment is, and what it is not

In methodological terms, a need is the gap between the current situation and a reference point: a legal standard, a service norm, the national average, the situation in other communities, or what people themselves want. Without a reference point there is no need, only a description. Knowing that a village has 40 children with disabilities says nothing about need until you know how many of them can access the services they are entitled to.

The specialist literature distinguishes four kinds of need, and a good assessment looks at all of them:

• Normative need: defined by experts or standards (for example, the maximum distance to a health post, or kindergarten places relative to the number of pre-school children).
• Felt need: what people say they lack when asked.
• Expressed need: what they actually demand, visible in waiting lists, applications and complaints.
• Comparative need: what a community lacks compared with similar ones.

The four do not coincide. People may not feel a need that standards clearly identify (such as missing preventive screening), or may press hard for something no indicator reflects. Comparing them is precisely where the value of the analysis lies.

3. When it is the right choice, and when it is not

A needs assessment is the right choice before an allocation decision: a local development plan, a social services strategy, the design of a project, or the selection of communities for a programme. It matters most when the options compete with one another and the decision must be justified publicly or to a funder.

It is the wrong tool if the intervention has already been decided and only confirmation is sought; the research then becomes a legitimising exercise, and communities notice that quickly. Nor does it replace a baseline for impact evaluation: if the programme is chosen and the aim is to measure its effect, the right design is the one described in our article on baseline, endline and impact evaluation. And if the main question concerns what people know, believe and do about a topic (health, finance, safety), a KAP study is more precise.

4. Designing a needs assessment: four sources of evidence

Existing data

The first stage is always an analysis of what already exists: official statistics, administrative records from the town hall, schools, health centres and social assistance, and reports from earlier projects. They supply normative and comparative need and show what does not need to be asked again in the field. The approach is described on our data analysis and desk research page. In small communities their limit is granularity: many figures do not exist at village level or are out of date.

The household survey

The survey measures how widespread problems are: what share of households have a member with a chronic illness who is not regularly treated, how many lack safe drinking water, how many children travel to another village for school. The sample is drawn at random from the community's households, stratified by settlement when the assessment covers several villages, and interviews are usually face to face on tablets, because the populations concerned often include older people and households without internet access. More on the method is on our social surveys page. The unit of analysis must be clear: some questions concern the household, others each of its members.

Key informant interviews

The mayor, the social worker, the family doctor, head teachers, the priest and leaders of local associations know about problems a survey cannot capture, especially those of small groups. In-depth interviews with 8-15 informants provide context, explanations and a picture of existing resources. Their risk is that they reflect the view of people who already have a voice.

Group discussions with residents

Separate discussion groups by category (young people, parents, older people, women, people with disabilities, minorities) show how problems are lived and which solutions would be accepted. Separation matters: in a mixed group, older residents do not contradict the mayor and teenagers do not speak up in front of their parents. For groups who do not come to meetings, interviews take place at home; strategies for these situations are covered in our article on hard-to-reach populations.

5. Who needs to be heard

The most common mistake in participatory assessments is confusing the people in the room with the community. Public meetings attract those who are mobile, informed and used to speaking. They do not attract mothers of small children, older people who cannot travel, seasonal workers or families living on the edge of the village. If priorities are set only from those present, they will serve exactly the groups that already had access to decision-making.

Two rules reduce the risk. First, a random household survey reaches people who take no part in public life. Second, every result is disaggregated by sex, age, disability, settlement and, where relevant, language, and small subgroups receive additional qualitative attention. The principles are set out in our article on gender-sensitive and inclusive research.

6. Analysis: from a list of problems to priorities

Evidence from the four sources is triangulated: a problem confirmed by statistics, the survey and informants is solid; one flagged by a single source is checked further. The result is a list of needs, each described in the same terms: who is affected, how many, how seriously, what services exist and what is missing.

Prioritisation then follows explicit criteria, agreed before the results are seen and together with those who will make the decision:

• Prevalence: what share of the population is affected.
• Severity: the consequences for health, safety, income or education.
• Vulnerability: whether the groups with the fewest resources are particularly affected.
• Feasibility: whether the problem can be solved with the resources and skills available.
• Community preference: how residents themselves rank the problems.

Each need is scored on each criterion, and the weight given to each criterion is discussed openly. The result is validated at a workshop with community representatives and local authorities, where the evidence is presented and the ranking discussed. Validation does not mean putting the results to a vote; it means checking that nothing important is missing and that the interpretation is sound.

7. How results become decisions

The final report does not end with a list of priorities but with their translation into concrete decisions. For each priority need it specifies three things: which intervention would reduce it (a new service, an extension of an existing one, an infrastructure investment, information), who can deliver it (the town hall, the district, an NGO, an external funder), and which indicator will show, in a year or two, whether the situation has changed.

Take a hypothetical case. The survey shows that a large share of older people living alone have not seen a doctor in the past year, and interviews with the family doctor and the social worker explain that the problem is transport, not the absence of a doctor. The decision is then not to build a health post, but to organise regular transport or home visits. Same need, two solutions with very different costs. Without combining sources, the choice would have been made on instinct.

For local authorities, the results feed directly into the development plan and the annual budget; for an NGO or a funded programme, into the project logic and its outcome indicators.

8. Limits and common mistakes

• The wish list: "would you like..." questions produce near-unanimous agreement and do not separate priorities.
• Elite capture: priorities reflect the interests of those who organise or dominate the consultation.
• Samples too small per settlement: in a multi-village assessment, village-level results need a minimum number of interviews, otherwise differences between villages are noise.
• Raised expectations: people asked about their needs expect them to be met. Clear communication of the purpose and the budget limits prevents frustration.
• A report with no follow-through: a needs assessment is only worth doing if it feeds directly into the plan, the budget or the project design, and the community learns what was decided and why.
• No baseline values: without clear starting figures, nobody will later be able to show whether the intervention reduced the need.

9. What each method contributes

10. What to ask the agency

• Which standards or reference points each need is defined against.
• Which existing data were analysed before fieldwork.
• How the household sample is drawn and how many interviews are planned per settlement.
• Which key informants are interviewed and how they are chosen.
• How vulnerable groups and people who do not attend public meetings are included.
• Disaggregation of results by sex, age, disability and settlement.
• The prioritisation criteria and who sets them.
• How results are validated with the community and with local authorities.
• The baseline indicators that will allow progress to be measured.

A well-run needs assessment does not decide on behalf of local authorities or the funder, but it makes the decision transparent: it shows what was chosen, on what basis, and what is left for the next stage. If you are preparing a local plan or a project, our social research team can help you design the needs assessment so that every voice counts.

See also: Social and political surveys Data analysis and desk research Focus groups and IDIs The KAP study Baseline, endline and impact evaluation Hard-to-reach populations

The Perspicuus research team · 29 July 2025