Indigenous Knowledge, Lived Experience, and Whose Evidence Counts
In South Africa, we are surrounded by evidence of progress. Strategies are adopted, budgets allocated, programs reach beneficiaries, infrastructure is completed, and annual reports record performance against targets. Yet communities describe a different reality.
A women’s economic empowerment program can report thousands of participants trained while women still lack access to land, finance, markets and the time required to build sustainable livelihoods because unpaid care work remains unequally distributed.
A water or sanitation project can be recorded as complete while rural households continue to walk long distances for water because taps are dry, infrastructure is poorly maintained, or supply is unreliable.
An environmental program can report hectares restored, trees planted or communities consulted without establishing whether ecosystems are recovering, local livelihoods are more secure, or communities have meaningful influence over natural resources.
A rural road can be counted as delivered while becoming impassable after heavy rain, cutting people off from schools, clinics and markets.
Taken together, these examples reveal a common pattern. Interventions may meet their formal targets without changing the conditions that made it necessary.
This contradiction sits at the heart of much of my work in monitoring, evaluation, research, learning and accountability. It has led me repeatedly to one question that sounds simple but is deeply political: who gets to define progress? Evaluation often begins with an approved plan. We examine whether activities were implemented, outputs produced and outcomes achieved, and rightly so. Public institutions and development organizations must account for resources, commitments and results. Yet an evaluation can confirm that a program performed according to plan without establishing whether people’s lives changed in ways they recognize as meaningful.
What Counts as Evidence?
When what is measurable becomes more important than what matters, institutional evidence can begin to outweigh lived experience of how progress is understood. Evaluation findings may show that a program has achieved its intended results, while the people affected by it describe a different reality. The question is not only whether evidence exists, but also which kinds of evidence are recognized as credible and whose perspectives shape the conclusions ultimately drawn.
I do not believe the answer is to reject formal indicators, administrative data or technical evaluation methods. The answer is to place them in conversation with other ways of knowing.
Indigenous and local knowledge systems do more than add local color or contextual detail to evaluation. They can challenge how change has been defined and what is accepted as evidence of progress. By Indigenous and local knowledge, I do not mean knowledge that fixed, romantic or untouched by science, technology and modern life. Rather, I am referring to locally rooted ways through which people observe patterns, interpret disruption, remember earlier conditions and decide how to respond. This knowledge may be carried through language, oral history, collective memory, relationships, repeated practice and close attention to social and ecological change.
It is also important not to speak of “the community” as though it shares a single, undifferentiated body of knowledge. An older farmer may interpret changes in rainfall through decades of planting experience. A woman responsible for food production and household care may notice that a new agricultural practice increases the time needed to collect water, prepare food, or manage crops. Young people may recognize changes in technology, markets or aspirations that older generations interpret differently. These perspectives may overlap, while also revealing different costs and benefits within the same intervention.
What Communities See That Indicators Miss
In rural and agricultural settings, a program may measure the number of farmers trained, inputs distributed, or hectares cultivated. These indicators capture important aspects of implementation and performance. They do not necessarily tell us whether a seed performs in local soil across several seasons, whether a water source is becoming less reliable, whether households have food during the period between harvests, or whether a farming practice can be sustained after funding and technical support end. Nor do they always reveal whether apparent gains in productivity have been achieved by transferring additional unpaid labor to women.
Communities may recognize change through signals that never appear in a results framework. Farmers may observe changes in rainfall patterns through years of planting experience, shifts in insect populations or vegetation, the performance of particular seeds, the condition of grazing land, or how long water remains available after rain. Individually, these observations may appear anecdotal. Considered together and over time, they can reveal patterns that a short project cycle or periodic survey may miss.
Indigenous and local knowledge offer more than an additional source of evidence. They may show that an evaluation is looking at the wrong period, asking the wrong people, or treating the wrong outcome as evidence of success. A program assessed immediately after implementation may appear effective, yet communities may judge it against their experience of dry seasons, failed harvests, earlier interventions and their ability to adapt without external support.
This knowledge is not a substitute for scientific, administrative or program evidence. Neither should it be accepted without examination simply because it is local. Its value lies in bringing different evidence systems into conversation. It can deepen formal evidence, question its assumptions, identify early signs of change and sometimes lead to different conclusions than conventional indicators alone.
From Evidence Collection to Evidence Use
These questions also shape how evaluations are used and the influence they ultimately have. We often assume that an evaluation will be used because the report is rigorous, the recommendations are clear, and the presentation is polished. In practice, evidence is more likely to influence decisions when the people expected to use it recognize their questions, realities and knowledge within it.
For me, this does not require inventing an entirely new form of evaluation. Participatory, reflective and use-oriented approaches are already well established within the field. The challenge is to apply these practices more consistently, ensure they have greater influence on decision-making, and prevent them from becoming procedural exercises.
Three Shifts Towards Better Evaluation
What would this look like in practice? I believe it requires three shifts in how we design, interpret, and use evaluation. The first is to define progress with people, rather than only for them.
First, progress needs to be defined with people, rather than only for them. Stakeholder participation is common, but it often begins after program outcomes, evaluation questions, and indicators have already been selected. Communities may be invited to provide information without having had meaningful influence over what the evaluation regards as success. Their involvement should begin early enough to shape the changes being examined and the evidence that would convince them that an intervention is working. Community-defined measures can complement formal indicators, expose missing dimensions of change and reveal harm or inequalities hidden by averages.
Second, interpretation should be treated as part of evidence generation rather than merely as a final validation step.. Evaluation teams, program managers and communities may interpret the same result differently because they experience the intervention from different positions. Structured reflection and validation should create room not only to confirm findings but also to examine disagreement, context, and competing explanations. A difference in interpretation is not necessarily an obstacle to be resolved; it may itself be an important finding about how power, benefit and progress are experienced.
Third, evaluation should be connected more deliberately to decision-making. Many evaluations identify intended users and produce recommendations, yet evidence use is still too often inferred from the completion of a report, a dissemination workshop, or management’s formal response. A report is not evidence of use. Use happens when findings change a budget, reshape an intervention, challenge an assumption, strengthen community action or lead an institution to stop doing something that is not working. This requires understanding what different users need to decide, producing evidence when those decisions are being made, and tracking what happens after recommendations are accepted.
These shifts do not make evaluation less rigorous. They strengthen it by drawing on a broader range of evidence and perspectives. Triangulation is strengthened by bringing together administrative records, quantitative results, qualitative accounts, community knowledge, and contextual observation. Evaluation becomes more relevant when the definition of success reflects the people whose lives are being evaluated, and more influential when evidence is generated through relationships rather than extracted and returned at the end of the process.
Progress should not exist only in a report. It should be visible in institutions, relationships, environments and everyday life. Most importantly, it should be recognizable to the people development is intended to serve.
When communities experience reality differently from what institutions report, evaluation should not explain the contradiction away. It should help us understand it.