Most social impact assessments do their real damage after they are approved. The document goes to the regulator, the permit is granted, and the report goes into a drawer. The impacts it predicted then unfold on the ground with no one systematically watching, because the assessment was treated as a gate to pass rather than a plan to run. The instrument that is supposed to prevent this is the Social Impact Management Plan, the SIMP, and in most projects it is the weakest link in the whole chain.
That is a failure of use, not of design. A SIMP, in the Queensland Government’s definition, is the plan that provides the practical basis for implementing the measures identified to manage a project’s social impacts. It is meant to be alive. It should say what will be monitored, by whom, how often, and what happens when the numbers move in the wrong direction. Treated that way, it is the difference between a project that catches a problem at month three and one that discovers it during a blockade. This article is about making the SIMP a working instrument inside a broader management system. Monitoring that both sides trust is what keeps a project and its neighbours in a working relationship over the long haul.
The SIMP is a component, not a standalone document
The first mistake is treating the SIMP as if it lived alone. It does not. It sits inside an Environmental and Social Management System, the ESMS. Under IFC Performance Standard 1, the ESMS is an iterative process of reviewing, correcting, and improving how a project manages its risks. The ESMS has nine components, and the SIMP is one action plan operating within them.
Those nine components are worth knowing because they show what a SIMP leans on. There is the policy that states the commitment. The identification of risks and impacts that feeds the plan. The management programs, where the SIMP itself sits. Organizational capacity and competency, meaning trained people with real budgets. Emergency preparedness and response. Stakeholder engagement. External communications and grievance mechanisms. Ongoing reporting to affected communities. And monitoring and review, which checks the whole system and lets management adjust.
Read that list and the failure mode becomes obvious. A SIMP written as a document, then filed, is a management program with none of the other eight components wrapped around it. No one owns it, because organizational capacity was never assigned. No signal reaches it, because the grievance mechanism was not wired into it. Nothing changes, because monitoring and review was a line in the report rather than a running process. The plan is not wrong. It is unplugged.
A working SIMP is plugged in. Its monitoring feeds the grievance data back into management decisions. Its commitments are owned by named people with resources to deliver them. Its reporting reaches the communities whose lives the impacts touch. The same discipline makes a grievance mechanism actually resolve conflict rather than just log it. The instrument only works when it is connected to the decisions it is supposed to inform.
What a monitoring plan actually contains
A monitoring plan is not a promise to keep an eye on things. It is a table, and the columns are the discipline. For each predicted impact, a well-built plan names the desired target or outcome, the monitoring method, and the party responsible. It sets the timing and frequency. It fixes the KPI that says whether the measure is working.
Fill those columns honestly and the vagueness has nowhere to hide. Take in-migration driving up population density near a site. The target might be a district masterplan that zones for the new demand on housing and utilities. The method is meetings with district planners. The responsibility is shared between the site’s government relations manager and the planning department, named, not “the company.” The frequency is quarterly. The KPI is a completed masterplan that addresses the increased demand. Now take economic displacement of farming households. The target could be a set share of producers adopting improved agricultural practices in the new season. The method is yield surveys and interviews. Extension officers monitor monthly, an implementation partner half-yearly. Different impact, same rigour.
Two things make this more than paperwork. First, shared responsibility. The strongest monitoring plans assign impacts jointly, a company manager alongside a government service or an independent partner, so no single party both causes and grades its own homework. Where air quality matters, the credible design puts the study in the hands of an independent third party, not the site. Second, the plan has to distinguish what it is measuring. Counting boreholes drilled is an output. Whether households actually have reliable water years later is an outcome, and outcomes are the honest measure of whether social performance worked. A monitoring plan heavy on outputs and light on outcomes is measuring effort, not effect.
Indicators both sides will trust
Indicators are where monitoring earns or loses credibility. A company can pick indicators that always look good. A community can suspect exactly that. The way through is to be deliberate about what makes an indicator sound, and to accept that not every indicator should be a number.
The familiar test is SMART: specific, measurable, attainable, relevant, and time-based. It suits the quantitative side, tonnes of dust, litres of water, weeks to close a grievance. But a lot of what matters in social performance is not captured by a clean metric, and forcing it into one distorts it. That is where SPICED comes in, a framework built for qualitative and participatory evaluation. Its indicators are subjective, meaning they value the unique perspective of the people affected. Participatory, developed with the community rather than for it. Interpreted and communicable, so they mean something to different audiences. Cross-checked and compared across methods and informants. Empowering, in that measuring them prompts reflection. And diverse and disaggregated, breaking data down so the experience of women and marginalized groups does not vanish into an average.
The point is not to choose one over the other. It is to use both, and to build the indicators together with the people whose reality they claim to describe. An indicator the company designed alone is an indicator the community has no reason to believe. The same logic runs through tracking social performance indicators as a measure of relationship quality: the value of the metric depends on both sides accepting it as fair. There are working models. The Indigenous Navigator, developed in 2017 by a consortium including the ILO and the Danish Institute for Human Rights, uses community questionnaires across ten domains of rights. Indigenous communities assess implementation on the ground themselves. The gap between what national law recognizes and what the community actually experiences becomes visible, and comparable over time.
Monitoring that changes the plan
Monitoring is worthless if nothing acts on it. This is the whole point of adaptive management, and it is where a live SIMP separates from a dead one. IFC PS1 builds the requirement in: ongoing monitoring exists so the management system can be adapted to real conditions during execution. The data is supposed to change the plan.
A documented example makes it concrete. On a mining project in the Democratic Republic of Congo, malaria was identified during impact assessment as a major risk. It threatened workers, their families, and surrounding communities across a licence area of scattered villages and two towns. The company built an integrated control program. More importantly, it built a monitoring system running on entomological and epidemiological data. The system collected malaria figures from the workforce, families, and the wider health zone, with twice-yearly surveys of schoolchildren. All of it was triangulated to track the program. And it was designed so that increases in malaria cases triggered changes: a resistance signal led to an insecticide change, a plateau led to a second bed-net campaign. The monitoring was not a report. It was a control loop.
That is what regulators mean, at their best, by adaptive management, and what a SIMP should institutionalize. The mechanism has three moving parts. A trigger, the indicator crossing a threshold that says something is off. A decision point, a named forum on a set cadence where the data is actually read, not filed. And a response, a defined action the trigger sets in motion. Most SIMPs have the first, occasionally the third, and almost never the second. Without a decision point, monitoring produces information that dies in an inbox. The connection to early problem detection is direct. It is the same thinking behind early warning systems for mining-community conflict. A signal only prevents an escalation if someone is obliged to look at it and empowered to act.
Making commitments survive the handoffs
The gap between a SIMP on paper and a SIMP in practice usually opens at the handoffs. The team that wrote the assessment is not the team that runs operations. The company that made the commitment often subcontracts the work that delivers it. Commitments made in year one get lost by year four, not through bad faith but through turnover, reorganization, and the simple fact that no one is holding the list.
The practical fix is a commitment register, and mining practice has converged on it. A public commitment register is a live document, examined across all project phases and shared with stakeholders and contractors, that tracks every social commitment made through the assessment. Alufer Mining publishes one. The register does three things a filed SIMP cannot. It survives staff changes, because the obligation lives in the document, not in someone’s memory. It binds contractors, because their scope references it. And it lets communities hold the project to its own word, because they can see the list. A commitment that is written down, owned, and visible is a commitment with a chance of being kept.
This matters most as the years pass and attention drifts. Impacts and the relationships around them do not stop when construction ends. The monitoring, and the accountability behind it, has to run for the life of the project and into closure. That is precisely when the original promises are hardest to remember and easiest to quietly drop.
Audit your SIMP before it goes in the drawer
Before you sign off a SIMP, run it through the SIMP Monitoring System Audit. It is a checklist of 17 checkpoints across five sections: system integration, monitoring plan quality, indicator credibility, the adaptive-management loop, and commitment accountability. It asks the questions that decide whether the plan will work. Does every impact have a named owner and a monitoring frequency? Were the indicators built with the community and disaggregated by gender? Is there a real decision point where a trigger forces a response? Does a commitment register bind your contractors? Score each item In place, Partial, or Absent. A run of Absents in the adaptive-management section means you have a reporting exercise, not a management system. Download the SIMP Monitoring System Audit, which shows you exactly which connection to wire before the document is filed and forgotten.
From document to living instrument
A SIMP is only as good as the system it lives in and the discipline that keeps it running. The plan itself is the easy part. The hard part is the plumbing around it: the owners, the triggers, the decision points, the register, the indicators both sides accept. Get that right and the SIMP stops being a compliance artefact. It becomes what it was meant to be, an early-warning and course-correction instrument that protects livelihoods and investments at once. A problem caught in monitoring is a problem solved before it becomes a dispute.
When monitoring does surface a problem, the response should be structured dialogue, not a defensive memo. A trend line both sides trust is the most useful thing you can bring to a mediated conversation. It replaces argument about what is happening with a shared fact about what is happening. This is where monitoring connects to the Social Accord Architecture, Thomas Gaultier’s field-tested framework for turning distrust into working relationships. Its closing phase, the Resilience Handover, gradually transitions management of the arrangement to local stakeholders so it becomes self-sustaining. The principle behind it is plain: a process depending on external facilitators indefinitely has not built infrastructure, it has built dependency. A monitoring system co-owned by the community is what makes that handover possible, because the community can only take over what it can see and trust. To design monitoring that holds a working relationship together over the life of a project, reach Thomas at thomas@thomasgaultier.com. A SIMP that no one reads protects no one. A SIMP that drives decisions protects everyone at the table.