SustIdx flagship methodology

Causal Ecology: what a company depends on, and what it’s doing to it.

Not another ESG score. Causal Ecology traces how much a business depends on a specific ecosystem service, whether that service is deteriorating, and — only when the evidence supports it — how much the business is contributing, and when that could return as financial risk.

See it on a company →Compare against public ESG scores

Dependency · Ecosystem state · Causal contribution · Feedback horizon

Why “Causal Ecology”

Most sustainability data, including SustIdx’s own Sustainability Signals, is built around carbon and disclosure — the right lens for a diffuse global problem, the wrong one for a local, specific one: a bottling plant on a stressed aquifer, a facility against a shrinking wetland. Causal Ecology covers freshwater and land/biodiversity dependency instead, resolved down to whether a real causal chain exists.

Informed by the research behind Beyond Carbon — Rebuilding Ecological Signals for Business and Finance, but built and run independently by SustIdx.

Three different things, often confused for one

A generic ESG score, Sustainability Signals, and Causal Ecology answer different questions — reading them side by side is the fastest way to see why Causal Ecology exists.

Dimension	Generic public ESG score	SustIdx Sustainability Signals	SustIdx Causal Ecology
What it measures	One blended score across unrelated topics.	Direction across nine topic families, from sourced news.	Dependency, ecological condition, exposure and causal contribution — kept separate, never averaged.
Causal specificity	None disclosed.	Event-level — every score traces to a matched story.	Locally attributable evidence is distinguished from diffuse, global mechanisms before anything is scored.
When data is thin	Still publishes a precise-looking number.	Confidence published; low-coverage issuers flagged.	Score withheld until evidence clears quality thresholds — never a fabricated figure.
Confidence disclosure	Rarely published.	Published per topic and issuer.	Shown alongside every output, never folded into it.
Update cadence	Often annual.	Recomputed hourly with the Universe Pool.	Recomputed alongside every Sustainability Signals refresh.
Transparency	Proprietary, undisclosed.	Full formulas and sources published.	Concepts, evidence standards and company-level provenance are public; the underlying equations and calibration are SustIdx's proprietary technical methodology.

Causal Ecology is never described as an ESG rating or a measure of total environmental harm — it’s a structured, evidence-gated exposure signal.

Four questions, asked in order

Causal Ecology never collapses these into one figure — that would hide the distinction that makes it useful: is a business exposed to ecological deterioration, or causing it? Those are different questions, and a company can answer them very differently.

What does the company depend on?

How operationally critical a given ecosystem service is to this business — how hard it is to substitute, how much of the business is exposed, how concentrated that exposure is geographically.

Is that ecological system under pressure?

Whether the service is actually deteriorating, from real, sourced evidence — not assumed, and not inferred from sector alone.

What exposure does that create?

Where dependency meets real deterioration, Causal Ecology publishes an exposure reading — how much business risk sits at that intersection, regardless of who's responsible for the deterioration itself.

Is the company contributing to it?

Only where the evidence supports a specific chain back to this company's own activity — deliberately the hardest bar to clear. Most companies won't clear it yet, and Causal Ecology says so explicitly rather than guessing.

Where exposure publishes, SustIdx also estimates — as a range, not a single date — roughly when that pressure could plausibly turn into a financial consequence. That timing estimate stays experimental until the ecological evidence behind it is driven by real ecological data rather than news coverage alone.

Not all evidence is equal

Before publishing anything, Causal Ecology places every company on a five-level evidence ladder. Stronger evidence unlocks more of the signal; weaker evidence means some or all of it stays unpublished rather than estimated.

Facility-attributed

A specific company facility is directly tied to real environmental data. The strongest tier.

Facility context

A specific facility is identified, short of full attribution.

Operational / regional

Real, issuer-specific disclosure plus a real operating location, without a facility-level link.

Contextual

The company's business activity and operating location are matched against real local ecological conditions — informative context, not a claim about this company's own facility.

No qualifying evidence yet

Not enough real activity, location or ecological-condition evidence currently exists for this company.

Local versus diffuse

Evidence strength is one axis; whether the mechanism is even local to begin with is another. Some pressures — a factory on a shrinking aquifer — are locally attributable in principle. Others, like atmospheric carbon, mix globally and can’t honestly be pinned on one company; those stay in SustIdx’s broader Sustainability Signals instead. Some issues are genuinely both — deforestation’s local land effects alongside its global carbon effect — tracked as two separate signals, never blended into one.

Clearing a level isn’t enough on its own

Even strong evidence has to pass SustIdx’s own quality checks before a number publishes. Until then, you’ll see what’s already known — what the company depends on — without a precise exposure or contribution figure standing in for evidence that isn’t there yet.

What the outputs mean

Four things to read on a company page:

Dependency

A contextual estimate of how materially the company relies on an ecosystem service, based on its sector and business activity. Available for almost every company, even before issuer-specific evidence exists — it's an estimate of reliance, not a measurement of environmental impact.

Ecological exposure

Published only once real, sourced evidence shows the relevant ecosystem is actually deteriorating. Where it can't yet be calculated responsibly, it simply doesn't appear.

Causal contribution

Published only where evidence ties the deterioration back to this specific company's own activity — the hardest bar to clear. Most companies won't have this yet, and the page says so rather than guessing.

Evidence quality

Shown alongside every output as context, not as another competing score — how much, and how strong, the evidence behind a given reading is.

Why you’ll sometimes see “not available” instead of a number

A zero means SustIdx measured the company and found genuinely nothing there. “Not available” means the evidence to make that call responsibly doesn’t exist yet — a very different, and much more common, situation. Causal Ecology never substitutes one for the other.

A worked example, in outline

Take a beverage company drawing water from a stressed aquifer. Dependency reads high — water is a core, hard-to-substitute, concentrated input. The aquifer’s deterioration is real and sourced, so exposure publishes too. If — and only if — a specific finding ties this company’s own extraction to that stress, causal contribution also publishes; otherwise it stays unestablished rather than estimated. A company in the same sector without that specific local evidence would show the same high dependency, but exposure and contribution would stay unpublished — never shown as a fabricated low number standing in for evidence that isn’t there.

Available across the platform, and its limits

On every company page

Shown alongside — never in place of — the base Sustainability Signals section.

Browse companies →

As an Index Lab screening field

Dependency, exposure and evidence level are all screenable when building a rules-based index.

Open Index Lab →

Public overview, proprietary methodology

This page covers concepts, evidence standards and interpretation. The underlying equations, calibration and gating rules are SustIdx's proprietary technical methodology — see Institutional & research use below.

Technical methodology — institutional access

Limitations, stated plainly

—Coverage is still early — most companies currently sit at “no qualifying evidence yet” while SustIdx's evidence pipeline expands.
—Evidence comes from public disclosures, regulatory filings and news — it lags real-world conditions and favours well-disclosed companies.
—Feedback-horizon timing is model-implied, not yet a calibrated real-world probability.
—A high dependency reading is a contextual, sector-based estimate — not a measurement of this specific company's facilities, unless stronger evidence says otherwise.

Institutional & research use

Versioned model runs, point-in-time reconstruction, and full provenance on every token. Companies at the highest evidence tier are eligible for systematic strategy or index research — reach out via the footer to discuss access to the full technical methodology.