Most participatory monitoring programmes are participatory in the way a guided tour is participatory. The community is invited to watch. A committee is formed, a few residents are taken to the sampling points, photographs are taken for the sustainability report, and the samples go to the company’s laboratory as before. Within a year the committee has stopped meeting and the residents have concluded that their presence changed nothing. The company has a programme on paper and a community that trusts its data no more than it did.
Real participatory monitoring is different in kind, not degree. The community shares control over what is measured, where, when, by whom, and how the results are read and acted on. That is harder for a company to accept, because it means giving up the ownership of information that most environmental departments regard as theirs. It is also the only version that works. A programme the community does not control produces data the community does not believe. Data the community does not believe changes nothing about the fear, the complaints or the road. Research on water monitoring in mining regions places programmes on a spectrum, from company-run with community observers to community-led. It is the programmes near the shared-control end that earn credibility. This article is about designing a programme that lands there, and about the practical decisions that decide whether it does.
What the community has to control
Five decisions determine whether a monitoring programme is trusted, and the community needs a real voice in each. First, what is measured. A programme that monitors the parameters in the permit and ignores the ones the community fears has answered the regulator’s question and not the village’s. The smell at dawn, the dust on the school roof, the fish that changed: those are the village’s parameters. Second, where and when. Compliance points on compliance schedules miss the events people observe; the programme needs points and moments the community chose. Third, who takes the sample or the reading. Trained community monitors, working alongside company or third-party staff, with both names on the record. Fourth, which laboratory, and who receives the results. A laboratory chosen together, independent of the company, reporting to both sides at once. Fifth, what happens when a result is bad. A response agreed in advance, so that the company cannot renegotiate it under pressure.
None of these is technically difficult. All of them require the company to treat the community as a party to the monitoring rather than an audience for it. The CAO’s guide to participatory water monitoring made this case years ago, and the evidence has only strengthened. A systematic review of citizen water monitoring finds gains in information, understanding, trust and conflict reduction wherever the community’s role was more than ceremonial.
Community monitors: selection, training, protection
The people who take the readings carry the programme’s credibility. Select them through the community’s own processes, from more than one settlement and more than one faction, with women among them. The monitors have to be trusted by the people they report to. Train them properly, in the field, by someone who is not the company’s environmental officer, in sampling, chain of custody, instrument use and record keeping. Pay them, modestly and transparently, because unpaid monitoring is abandoned within a season and paid monitoring is accused of being bought if the payment is hidden. And protect them. A monitor who reports an exceedance can be exposed to pressure from the company, from local leaders with an interest in the project, or from neighbours who depend on it. The programme should state, and the company should honour, that a monitor faces no consequence for reporting what the instruments show.
Work relayed to me by a practitioner in one of my trainings makes the point. A dust programme placed a simple deposit gauge on the roof of a village school and trained two teachers to empty and weigh it monthly with a company technician present. The gauge was crude next to the fence-line monitor. But the teachers’ figures were believed, and when they rose one dry season the company adjusted the watering schedule before a single complaint was filed. The technology mattered less than who read it.
The same logic applies to instruments. A community monitor with a handheld turbidity meter, a pH strip and a notebook produces data of lower precision than a fence-line station and higher credibility. Where precision matters, the answer is to pair the two: the community’s reading as the trigger, the laboratory’s as the confirmation. Research on community-based monitoring in Peru and Chile shows communities adopting monitoring technology on their own account when companies would not share it. That is a warning as much as an encouragement. Better that the instruments be shared than duplicated.
Chain of custody in front of everyone
Data is contested at its weakest link, and in participatory monitoring the weakest link is almost always custody. A company that wants to dismiss a bad result will question how the sample was handled. A community that wants to dismiss a good one will question who had it between the river and the laboratory. The answer is procedure done in the open. Sealing, labelling and logging happen at the sampling point in front of both parties. Duplicate samples go to two laboratories when a result matters. Transport is documented, and the cooler does not disappear into a company vehicle. The laboratory reports to both parties simultaneously, not to the company for review first. A programme that gets custody right has removed the argument that both sides reach for when the number is inconvenient. It costs a second cooler, a duplicate label and an hour of two people’s time. It saves the meeting where the result is thrown out because nobody can say who carried the bottle.
Reading and sharing results
Numbers do not speak for themselves, and the party that interprets them controls the conversation. The programme therefore needs an interpretation step that both sides own. The community’s own technical adviser, funded through the programme but answerable to the community, reads every result alongside the company’s specialist. Results are presented first in the community, at the water point or the school, in local languages and in visual form, before any written report circulates. Trends are shown, not only single readings, because a single clean result does not answer a fear and a single bad one does not prove a pattern. And the presentation includes what will be done, because a result without an action is a lecture.
Sharing has to be routine, not reactive. A monthly or quarterly rhythm, kept whether the results are good or bad, teaches the community that the data is theirs to see. Publication only when the news is good teaches the opposite.
There is a role for the regulator here too. Results shared with the regulator at the same time as the community, in the same form, close a gap. It is the gap in which the company is suspected of telling each a different story. Where the regulator is weak or distant, the joint programme may be the only monitoring anyone reads. It should be designed to be handed over to the public authority when that authority is ready to take it. The point I make in Building Trust Through Data Transparency applies directly: transparency that arrives only under pressure is read as concealment that failed.
The committee that keeps meeting
Most programmes are governed by a joint monitoring committee, and most committees die within two years. They die because they have no decisions to make, or because the company controls the agenda. They die because members are not paid for their time, or because the results they discuss never change anything. A committee that survives has a written mandate with real decisions in it. It approves the sampling plan, chooses the laboratory, receives results first and decides how they are communicated. It triggers the agreed response to an exceedance and reviews the programme annually. It has rotating membership from the affected settlements, a chair who is not a company employee, and an independent facilitator for the meetings where the results are contested. And it sits inside the wider engagement structure, so that what it learns reaches the forum where the relationship is governed. The social impact monitoring I describe in Social Impact Monitoring: The SIMP Inside a Management System and the environmental monitoring described here should report to the same place.
Design the programme before you need it
The Participatory Monitoring Design Roadmap sets out the build in five phases and 16 steps. The phases are scope and control agreed with the community, selecting and training monitors, custody and laboratory arrangements, interpretation and communication, and governing and sustaining the programme. Score each step In place, Partial or Absent, and read the result phase by phase. Absents in the first phase mean the community is being invited to watch rather than to decide, which is the failure that empties committees. Absents in the last mean the programme will not survive its second year. Use it to design a programme before a dispute, or to rebuild one that has become a tour. Download the Participatory Monitoring Design Roadmap
Shared data is the foundation of a working relationship
A company and a community that produce their environmental data together have something that neither can produce alone: a record both believe. That record protects the households, who no longer have to choose between fear and the company’s word. It protects the project, whose environmental performance is now defended by the people who would otherwise be its accusers. Livelihoods and investments rest on the same shared numbers.
Building that record is a negotiation about control, and it is best held in a structured process before the first bad result rather than after it. The Social Accord Architecture treats participatory monitoring as a core part of the Feedback Loop, designed in Blueprinting. The parameters, points, monitors, laboratory, adviser and response rules are written into the agreement as commitments. The joint committee reports to the council where company and community both sit, and a contested result goes to the mediation pathway rather than to competing press releases. Dual accountability holds the company to acting on the shared data and the community to raising concerns through the programme it helped design. Where a monitoring programme has already collapsed into mutual suspicion, an independent facilitator can usually rebuild it, starting with the one decision the community was never given: what to measure.
One thing to do this quarter. Ask your joint monitoring committee, if you have one, when it last made a decision the company then carried out. If nobody can remember, the committee is an audience. Give it the sampling plan to approve, and see what changes. In my experience the first thing that changes is attendance. To design a participatory monitoring programme or to rebuild one that has lost the community’s trust, write to me at thomas@thomasgaultier.com.