Methodology · BEF Impact Assessment Platform

The picture in one chart

Illustrated with the Climate objective and the worked methane-satellite example. Edit any goal’s shares on the Wedge Weights page and this redraws.

a binding constraint (height = its wedge share of the gap) a constraint this project works on expected closure of that constraint expected impact — solid to P10, tick = mean P10–P90 uncertainty band

Why wedges

Four ultimate objectives anchor the Fund’s mission to protect and restore the natural world: holding 1.5 °C / net-zero CO₂, halting and reversing deforestation, halting human-induced extinctions, and clean air and energy access through a just transition. Everything else — methane, power systems, protected areas, restoration, finance — is a binding constraint beneath one of the four, not an end in itself.

Between where the world is realistically heading — the counterfactual — and an objective being met lies a gap. Each objective’s gap is divided into the binding constraints that hold it open; each constraint carries a wedge share of the gap: how much closer to the objective we get if it is fully removed. Shares sum to 100% per objective. A project’s job is to shrink one or more wedges.

The scoring chain

① Impact wedge

Σ of the wedge shares of the constraints the project works on — the slice of the gap it is even trying to close.

→

② Total projected impact

Σ (contribution% × wedge × plausibility%) across those constraints — points of the gap closed if things go to plan.

→

③ Enabling success factor

A Monte-Carlo estimate of the chance the external dependencies all hold — reported as a mean with a P10–P90 band.

→

④ Expected impact

TPI × the success factor — the goal-specific expected result, discounted for what lies outside the project’s control.

→

⑤ Global significance

Expected impact × the goal’s comparative rating% — a cross-goal score on the same 0–100 scale, comparable across goals and programs.

Significance — how goals compare

Each objective carries a comparative rating (0–100) built from four criteria: global impact magnitude (0.40), BEF strategic priority (0.30), urgency & irreversibility (0.20) and co-benefit breadth (0.10). Because global significance multiplies expected impact by this rating expressed as a percentage, it reads on the same 0–100 scale as the goal-specific number — so a biodiversity grant and a climate grant can be laid side by side. The goal dials at the top of every page let you set these ratings and watch the whole portfolio re-rank.

Uncertainty — the enabling success factor

Most projects only deliver if things outside their control hold — a policy is enforced, co-funding lands, an adoption rate rises. Rather than a crude min–max band, each enabling condition is treated as an uncertain probability (a Beta distribution around the entered chance), the conditions are combined by their logic (all-must-hold, any-one, or k-of-n), and the result is simulated to a distribution. We report the mean as the expected value and the P10–P90 as the band; ranking uses the mean. Across the 204-grant portfolio, the number of external dependencies each description names drives how much its impact is discounted.

What this is — and is not

Portfolio assessments here are produced by a transparent keyword engine from grant title and description — not a human review or a language model. Treat every figure as a first-pass draft to verify. Wedge shares, comparative ratings and constraint definitions are the framework’s illustrative starting values, to be calibrated by the Fund. The environmental & social safeguards gate is not auto-scored: a flag simply marks text that warrants a manual harm screen. A project’s ultimate goal often differs from the Fund program that houses it, which is why the Performance-by-Goal and Performance-by-Program breakdowns differ by design.