Sustain ESG Lab
Momentum, risk and change across 5,000+ listed companies — each classified by its ESG state.
What's moving
The biggest signal shifts in the last 30 days. Each row shows what changed, why, and what to do.
Board-level governance issue flagged. Incidents +45% vs last quarter.
3 quarters without improvement. Controversy 2.1× sector avg.
3Q positive momentum → upgraded from Improving to Leader.
Flat → Improving. Climate commitments tightening vs last quarter.
Biggest 12-month improvers
Top names gaining ESG score over the last 12 months.
| # | Company | Sector | ESG | State | 12m trend |
|---|---|---|---|---|---|
| 1 | Siemens AG SIE.DE | Industrials | 78 | Very strong | +14 |
| 2 | Schneider Electric SU.PA | Industrials | 74 | Very strong | +11 |
| 3 | Unilever plc ULVR.L | Consumer Staples | 71 | Very strong | +9 |
| 4 | AstraZeneca plc AZN.L | Healthcare | 69 | Strong | +8 |
| 5 | RELX plc REL.L | Media | 66 | Strong | +7 |
Free and open
All analytics published openly. No paywall, no login, no tracking.
Transparent by design
Every score, threshold, and limitation is documented. Nothing is a black box.
Built for decisions
Portfolio managers, researchers, and allocators. Transparent methodology, open access.
Where these signals come from
Our signal engine in 5 steps — entity resolution, event normalisation, cross-source validation.
Show
This platform does not overlay third-party ESG datasets. It runs a structured pipeline that resolves fragmented entity data, normalises inconsistent event signals, and validates across sources before any score is constructed. The value is in the issuer-level data engine — not the raw inputs.
Source ingestion
Event streams, climate commitment registries, issuer identity databases. No single source is treated as authoritative.
Entity resolution
Names, tickers, ISINs, LEIs reconciled into one issuer graph. 70,000+ records, duplicates collapsed.
Event normalisation
Incidents classified by severity and pillar. Duplicates deduplicated across sources. Temporal alignment.
Cross-source validation
Claims tested across sources. Commitments cross-referenced with incident patterns. Contradictions flagged.
Signal construction
Quality, Risk, Momentum layers computed independently on validated data. State classification and transition detection produce clear, evidence-backed outputs — not raw scores.
How to read this page
Every company sits in one of five level states, set by where its composite ESG score falls on the published bands.
Signals
What's moved in the last 30 days across 5,000+ listed companies — filtered, ranked, and action-tagged.
How to read state transitions
The five transition types that our pipeline tracks and what each signal means.
Show
Analytics
Leaders, risk & exclusions, momentum, exposure and portfolio impact — open any to explore the latest outputs.
ESG Leaders
01Top-scoring companies across the universe — the highest-conviction ESG positions.
Use when: building a positive-tilt portfolio.
Risk & Exclusion Monitor
02Companies with high controversy exposure — flagged for exclusion or close watch.
Use when: running exclusion screens.
Momentum Signals
03Who's getting better, who's getting worse — and how fast.
Use when: spotting changes before they price in.
Exposure Map
04ESG risk by sector and country — where exposure is concentrated.
Use when: making allocation decisions.
Portfolio Impact
05Portfolio ESG score, drivers, and what-if swaps — test substitutions before you make them.
Use when: stress-testing portfolio construction.
Momentum Signals
Who's getting better, who's getting worse — and how fast.
Biggest improvers (12m)
Showing 5 of 15| # | Company | Sector | ESG | State | 12m |
|---|---|---|---|---|---|
| 1 | Siemens AG | Industrials | 78 | Very strong | +14 |
| 2 | Schneider Electric | Industrials | 74 | Very strong | +11 |
| 3 | Unilever plc | Cons. Staples | 71 | Very strong | +9 |
| 4 | Vestas Wind Systems | Industrials | 70 | Very strong | +8 |
| 5 | AstraZeneca plc | Healthcare | 69 | Strong | +8 |
| 6 | ABB Ltd | Industrials | 68 | Very strong | +6 |
| 7 | RELX plc | Media | 66 | Strong | +7 |
| 8 | GSK plc | Healthcare | 64 | Strong | +6 |
| 9 | Nestlé | Cons. Staples | 62 | Strong | +5 |
| 10 | Ørsted | Utilities | 75 | Very strong | +5 |
| 11 | SAP SE | Technology | 67 | Strong | +5 |
| 12 | Novo Nordisk | Healthcare | 68 | Strong | +4 |
| 13 | ASML Holding | Technology | 65 | Strong | +4 |
| 14 | Linde plc | Materials | 63 | Strong | +3 |
| 15 | Iberdrola | Utilities | 72 | Very strong | +3 |
Biggest decliners (12m)
Showing 5 of 15| # | Company | Sector | ESG | State | 12m |
|---|---|---|---|---|---|
| 1 | BP plc | Energy | 28 | Very weak | −12 |
| 2 | Glencore plc | Materials | 31 | Very weak | −10 |
| 3 | Shell plc | Energy | 33 | Very weak | −9 |
| 4 | Rio Tinto | Materials | 36 | Weak | −7 |
| 5 | TotalEnergies | Energy | 34 | Very weak | −6 |
| 6 | ExxonMobil | Energy | 30 | Very weak | −5 |
| 7 | Chevron | Energy | 32 | Very weak | −5 |
| 8 | HSBC Holdings | Financials | 48 | Weak | −4 |
| 9 | Anglo American | Materials | 38 | Weak | −4 |
| 10 | Equinor | Energy | 45 | Weak | −3 |
| 11 | ArcelorMittal | Materials | 39 | Weak | −3 |
| 12 | Saudi Aramco | Energy | 29 | Very weak | −3 |
| 13 | Barclays | Financials | 50 | Weak | −2 |
| 14 | Petrobras | Energy | 35 | Very weak | −2 |
| 15 | Vale S.A. | Materials | 37 | Weak | −2 |
ESG Leaders
The highest-conviction ESG positions — top-scoring companies across the universe.
Top ESG leaders
Showing 5 of 15| # | Company | Sector | ESG | State | 12m |
|---|---|---|---|---|---|
| 1 | Siemens AG | Industrials | 78 | Very strong | +14 |
| 2 | Ørsted | Utilities | 75 | Very strong | +5 |
| 3 | Schneider Electric | Industrials | 74 | Very strong | +11 |
| 4 | Iberdrola | Utilities | 72 | Very strong | +3 |
| 5 | Unilever plc | Cons. Staples | 71 | Very strong | +9 |
| 6 | Vestas Wind Systems | Industrials | 70 | Very strong | +8 |
| 7 | AstraZeneca plc | Healthcare | 69 | Strong | +8 |
| 8 | ABB Ltd | Industrials | 68 | Very strong | +6 |
| 9 | Novo Nordisk | Healthcare | 68 | Strong | +4 |
| 10 | SAP SE | Technology | 67 | Strong | +5 |
| 11 | RELX plc | Media | 66 | Strong | +7 |
| 12 | ASML Holding | Technology | 65 | Strong | +4 |
| 13 | GSK plc | Healthcare | 64 | Strong | +6 |
| 14 | Linde plc | Materials | 63 | Strong | +3 |
| 15 | Nestlé | Cons. Staples | 62 | Strong | +5 |
Exposure Map
ESG risk by sector and country — where exposure is concentrated.
Sector risk direction
Highest-exposure sectors| # | Sector | Direction |
|---|---|---|
| 1 | Oil and Gas | → Mixed |
| 2 | Utilities | → Mixed |
| 3 | Media | → Mixed |
| 4 | Development Banks, Central Banks, and Export Credit Agencies | → Mixed |
| 5 | General Industrials | → Mixed |
| 6 | Construction and Materials | ↓ Deteriorating |
| 7 | Chemicals | ↓ Deteriorating |
| 8 | Alternative Energy | ↓ Deteriorating |
Concentration watchpoints
Sector & region| Type | Name | Signal |
|---|---|---|
| Sector | General Industrials | Elevated risk avg risk-quality 32, 44% excluded |
| Sector | Technology Hardware and Equipment | Elevated risk avg risk-quality 34, 27% excluded |
| Region | North Macedonia | Rising exposure 1% excluded, incident density elevated |
| Sector | Oil and Gas | Improving momentum avg Δ30d: +0 |
Portfolio check
Paste up to 50 tickers or upload a CSV. Get a composite ESG score and an ESG-state breakdown across your holdings.
—
—
—
Signal history — last 90 days
Methodology
How every number on this site is produced — score by score, column by column.
What this page covers
Every value shown on this site — the composite ESG score, the Level and 12m trend columns, incident counts, severity and persistence words, pillar drivers, the ESG states, and the portfolio check — is explained below in the order you meet it on the site. Where a display uses a shorthand, this page gives the full derivation.
1 · The composite ESG score (0–100)
Each covered company receives one continuous score from 0 to 100, built from four components:
Risk Quality reflects the company's controversy exposure — how many incidents it has, how severe they are, and how recent. Commitment reflects verifiable climate commitments (for example SBTi status), cross-checked against the incident record rather than taken at face value. Clean Record rewards a sustained absence of material incidents. Coverage Depth adjusts for how much independent data exists on the company, so a high score cannot be earned simply by being under-reported. There is no pass/fail step: the score is continuous, and the bands below are read off it.
2 · The three measures and their words
The site describes each company along three separate measures. Each measure has its own vocabulary, and one word belongs to exactly one measure — a page labels the measure it is about, and other measures show as bare numbers.
Level — how good the record is now. Driven by the composite score; band cuts sit at 65 / 57 / 50 / 43, the lower bound of each band inclusive. Recomputed quarterly.
| Score band | Level |
|---|---|
| 65–100 | Very strong |
| 57–64 | Strong |
| 50–56 | Moderate |
| 43–49 | Weak |
| below 43 | Very weak |
Trend — which way the record is moving over 30 days. Driven by momentum; band cuts sit at ±60 / ±20. Moves whenever incident data refreshes.
| Momentum | Trend |
|---|---|
| ≥ +60 | Strongly improving |
| +20 to +60 | Improving |
| −20 to +20 | Flat |
| −60 to −20 | Deteriorating |
| ≤ −60 | Strongly deteriorating |
Exclusion — whether the company trips a screen. Output of the exclusion screens: Critical / Severe / Elevated / Watch, in descending severity. Exclusion bands follow the screens that produce them.
State words describe the company's measured ESG position under these published thresholds. They are properties of the company in our methodology, not recommendations to any person.
3 · Sector caps
Composite scores in Oil & Gas and Mining are hard-capped at 45; Coal and Tobacco are capped at 35. Caps apply after all other components, regardless of how well an individual company scores on conduct. This is a deliberate product-level exclusion on what a company does, kept separate from the conduct engine that measures how it behaves.
4 · The 12m trend column
Wherever you see 12m trend, it is the change in the composite ESG score over the rolling twelve months, expressed in score points. It is computed from the same incident and commitment inputs as the score itself, so a company's Level and trend can legitimately disagree — a low score can be improving fast (a recovery signal, not a quality signal), and cards flag that divergence explicitly when it occurs.
5 · Incidents, severity, and persistence
Card evidence lines follow one template: Incidents: 12 → 1 · Severity: low · not recurring.
Incidents counts distinct ESG incidents attributed to the company over the trailing 24 months, after deduplication across sources; an arrow (12 → 1) compares the prior year with the most recent year. Severity aggregates per-incident severity and is shown as a word: low (≤10), moderate (11–25), high (26–39), critical (40+). Persistence measures how continuously incidents recur and is shown as: ongoing (persistence ≥ 0.25), easing (0.10–0.25), not recurring (< 0.10). The severity and persistence cut points are the current calibration and are reviewed each quarter.
6 · Signal drivers (Env / Soc / Gov)
Next to each signal you will see Env / Soc / Gov labels showing which ESG pillars that signal draws on. In the current release these come from the company's sector profile rather than from per-company incident attribution — a sector-level starting point, not a claim that every individual incident has been traced to a pillar. Per-company attribution is on the development roadmap, and this section will be updated when it ships.
7 · Quarterly sector call-outs
The five cards on the homepage (Very strong / Strong / Moderate / Weak / Very weak) are sector-level summaries for the current quarter. Each sector's constituent scores and 12-month trends are aggregated; the sector with the strongest positive momentum among adequately covered sectors appears under Very strong, the weakest under Very weak, and the sector whose aggregate score moved least appears under Moderate. The names quoted on each card are the largest individual contributors to that sector's move. If no sector qualifies for a state in a quarter, the card says so rather than forcing a pick.
8 · The portfolio check
The screener resolves each ticker against the covered universe. The portfolio composite is the weighted average of resolved holdings' ESG scores — weighted by position value (quantity × price) when a CSV provides them, equal-weighted otherwise. Unresolved names are listed separately and excluded from both the numerator and the denominator, so they can never silently drag the average. "Gap to Good / Strong" is the distance from your composite to the 57 and 65 band cuts. Exclusion-signal counts show how many holdings sit in the exclusion band; what-if swaps recompute the composite with flagged holdings removed.
9 · Sources and update cadence
Incident data comes from Event Registry, which supplies the 24-month incident history, severity, pillar classification and UN Global Compact flags, together with the live 30-day incident view. Climate commitment data comes from the SBTi registry. Emissions data used in sector, country and company analysis comes from Climate TRACE. Issuer identity comes from a reconciled security master covering names, tickers, ISINs and LEIs.
These sources update on different cadences, so the site makes no single global "last updated" claim. Composite scores and the level words read off them are recomputed quarterly. Trend is a 30-day measure and moves whenever incident data refreshes. Exclusion bands follow the screens that produce them.
Limitations
Incident data carries media-coverage bias: well-covered companies accumulate more recorded incidents than poorly covered ones with similar conduct. Scores reflect reported controversy exposure and verifiable commitments, not a complete picture of ESG performance. Sector-level pillar tags are a prior, not per-incident attribution. Coverage varies by region and company size. Everything published here is research — a description of how companies measure under a stated methodology. None of it is investment advice or a personal recommendation to any person.
What is Sustain ESG Lab
An early-stage analytics platform that organises fragmented ESG data into clear, checkable analysis for investors, researchers, and wealth managers.
What we're building
ESG information is scattered across sources, inconsistent between them, and slow to reach the people who could actually use it. We collect it, reconcile it, check claims against each other, and publish the result in a clear, checkable form — with the working shown. We publish analysis, not recommendations.
Where we're headed
Today, good ESG information mostly sits inside large institutions. We want anyone — from a first-time investor to a professional analyst — to be able to check the ESG record of the companies they hold, quickly and for free. We cover 5,000+ listed companies today, and plan to add physical climate risk and private markets over time.
Who this is for
Most individual investors don't have time to research the sustainability of every company they hold — but the interest is clearly there. Research cited by the UN Global Compact (drawing on Morgan Stanley, 2025) finds that 88% of investors are interested in portfolios that integrate sustainability, and academic work suggests around two-thirds of millennials want their investments to reflect their social and environmental values (Hildebrand et al., 2020; Pelster et al., 2024). The same research argues the case is financial as well as ethical.
We built this site to close the gap between that interest and the information: to make a company's ESG record easy to reach and easy to check, for anyone. Professional analysts and first-time investors see the same evidence, described the same way. And the boundary is deliberate — we describe companies under a published methodology; we don't recommend investments.
Operating principles
Raw data sources have known biases. We normalise, validate, and reconcile across sources before publishing anything.
Every company state carries a single level label — Very strong / Strong / Moderate / Weak / Very weak — set by where its composite ESG score sits on the published bands. One vocabulary, the same for everyone.
Score thresholds, sector caps, and time horizons are documented and stable. We publish how we score before we tell you what we score.
Media-coverage bias, sector-specific data gaps, and the boundary between an exclusion screen and a regulatory fund label — every known limitation is documented alongside the scores it affects.
Talk to us
For institutional access, research collaborations, or data partnerships:
About us
Who is behind Sustain ESG Lab, and why it exists.
Why this exists
Sustain ESG Lab grew out of a simple observation: investment choices carry real environmental weight. They fund the companies doing serious environmental damage, and they fund the ones doing better — and most people never see the information that would let them tell the difference. That information exists, but it sits behind paywalls, jargon and institutional tools.
This site shows how listed companies measure up on environmental, social and governance evidence — scores, incidents, commitments, and how they change over time. We bring together established data sources and present what they show in an accessible, checkable form.
Sustain ESG Lab is not authorised to give financial advice, and nothing here is an investment recommendation. What we publish is research: information people can weigh in their own decisions.
Who we are
Dharma Couture — Founder. Final-year BSc Environment and Sustainable Development with Economics student at the London School of Economics.