Method
How we score, in outline. The full methodology, version 1.0, is published with the launch at COP17 in October 2026, and versioned from then on.
Four scores, one composite
Every company we cover receives four scores from 0 to 100, each relative to its sector and in absolute terms.
| Score | What it measures | Built from |
|---|---|---|
| Dependency | How much revenue relies on ecosystem services: water, pollination, soil, climate regulation, raw materials | ENCORE dependency ratings, revenue by segment and geography, the condition of the ecosystems at the company's locations |
| Impact | The pressure the company exerts: land-use change, water use, pollution, over-exploitation, invasive species, greenhouse gases as a nature pressure | ENCORE impact ratings, operational footprint, asset-level overlays such as proximity to Key Biodiversity Areas and deforestation alerts in sourcing regions |
| Transparency | The quality of nature disclosure, scored against the TNFD's fourteen recommended disclosures and the ISSB's proposals | Company reports, read and scored by Nature Analyst with human review; a bonus for assurance and for location-specific disclosure |
| Response | Evidence of action: targets, spend, supply-chain traceability, restoration commitments, certifications, litigation exposure | Company disclosures, SBTN target lists, certification registers, court records |
The BE Score weights them: Impact 35%, Dependency 25%, Transparency 20%, Response 20%. Every score is published with a confidence band from A to D that reflects how much data sits behind it.
The Extinction Discount
The Extinction Discount converts the scores into money: the share of a company's enterprise value exposed to nature loss under defined scenarios, split into what is already priced by the market and what is not. The first scenarios are the 2026 to 2027 El Niño, policy tightening under the Global Biodiversity Framework by 2030, pollinator and soil degradation by 2035, and water-basin stress by 2030. Each carries a stated probability and a sensitivity table. The model is published as a working paper, open to academic critique, and versioned.
The Nature Quadrant
Dependency on one axis, impact on the other, bubble size for market capitalisation, colour for transparency. Four regions with plain-English names. A fund appears as a cloud of its holdings; the fund's centroid is its position.
Fund Truth
Every fund and ETF with "biodiversity", "nature" or "natural capital" in its name or objectives is scored on the aggregate impact, dependency, transparency and response of its holdings, and on how much of its revenue comes from verified nature-positive activity rather than from industrials with a tenuous link. Funds receive a Truth Score from 0 to 100 and one of five bands: Regenerative, Aligned, Adjacent, Labelled, Contradictory.
Governance
- Right of reply. Companies and funds receive ten working days' notice of a first public score, with the data behind it. Corrections are logged on the Corrections page.
- Conflicts. Advisory clients are scored by the same pipeline as everyone else and reviewed by someone not on the engagement.
- Sources. No figure is published without a source record. The scoring weights and the pipeline are trade secrets; the logic is public.
- Data ethics. No personal data. Supplier and plot data supplied by clients never feeds public scores without consent.
- Review. An annual external review of the methodology by an academic partner, starting in 2027.