The commitment register: how projects lose track of what they promised

Ask a project team in year two to state every environmental and social commitment the project carries. On most projects, nobody can. Not because the team is careless, but because the commitments were never in one place: some came from the ESIA's mitigation measures, some from the RAP, some from conditions in the lender's agreement, and some were flowed down into the works contract by a lawyer who has since left.

Four documents, four authors, four moments in time, and no list. The commitments that were never transcribed into anyone's workplan are simply not delivered — and they resurface at completion audit, when the remedy is expensive and the people who could explain the omission have moved on.

What the register actually is

A Mitigation and Monitoring Commitment Register is the transcription. One row per commitment, each traceable back to the document that created it, each with a named owner, a stated means of verification and a monitoring frequency. It is not a new obligation; it is an inventory of obligations the project already has.[1]

The instrument is standard on Ugandan infrastructure projects in the form of a numbered commitment set — RAP01 through RAP47 — covering workforce, engagement, grievance handling, health and safety, national content, waste and reporting. The numbering is a local convention. The discipline it enforces is not.

Means of verification is where registers fail

The column that decides whether a register is useful is the one naming what a reviewer would ask to see. Write it specifically — a signed attendance sheet, a waste transfer note with a returned disposal confirmation, a payroll record — and the commitment becomes auditable. Write it as "monitoring reports" or "site records" and it does not.

This matters more than it sounds. An unauditable commitment is not a commitment with weak evidence; it is a commitment that can go unmet indefinitely without anybody noticing, because there is no artefact whose absence would be conspicuous.

Who owns it

The employer owns the register; the contractor owns most of the rows in it. That split is what makes the instrument work, and inverting it is the most common way to neutralise it. A register maintained by the party being assessed is a self-report with extra formatting.

In practice the contractor files evidence against each applicable commitment for the period, and the employer's ESHS team verifies and scores it. On projects that use the instrument seriously, that score then conditions payment certification — which is what turns a list into a control rather than a reporting exercise.[2]

Agree the scoring rule before the first period

Adherence is the proportion of applicable commitments met. The word doing the work is applicable. A commitment about hazardous waste disposal is not unmet during a period in which no hazardous waste arose; scoring it "not met" understates performance, and quietly dropping it from the denominator overstates performance in the other direction.

Decide the applicability test and the adherence threshold before the first assessed period. Where the score gates money, both will be argued about, and a rule settled after the first disputed invoice has already lost the authority it was created to carry.

What the register exposes on day one

Sort a completed register by the management plan each commitment belongs to. On most projects one or two plans will be carrying commitments that no existing document actually covers — obligations picked up from the works contract that were never written into a management plan at all.

That is the register earning its cost before a single period has been scored. Finding an uncovered obligation in month three is a drafting task. Finding it at completion audit is a finding.[3]