We started by misreading a footnote and never looked back

Three years ago, we confessed to each other that most ESG scoring felt like astrology with better branding. Since then, we have built a small team obsessed with cross-sectional signals, cleaner data pipelines, and the quiet thrill of a well-specified regression. We do not sell certainty. We model questions.

Evidence-first culture
Collaborative research

Say hello Our story

We believe ESG integration should be systematic, cross-sectional, and grounded in empirical evidence rather than box-ticking exercises

Question everything

We treat every ESG rating as a hypothesis, not a verdict. Our models start with scepticism and end with confidence intervals. We have learned that the most expensive mistake in sustainable investing is accepting vendor scores at face value without understanding the underlying construction methodology.

Cross-sectional focus

A single company's ESG profile means little in isolation. We compare signals across thousands of issuers simultaneously, controlling for sector, region, and market regime. This approach reveals patterns that firm-level analysis consistently misses.

Transparency first

We document every data choice, every imputation rule, and every model assumption. If we cannot explain a signal to a sceptical colleague, we do not use it. Intellectual honesty is our only non-negotiable constraint.

Practitioner relevance

Academic rigour without implementation utility is a hobby, not a service. We write code that runs in production environments and produce research that portfolio managers can actually act upon within their existing workflows.

Finding your analytical confidence at any career stage

Whether you are building your first quantitative ESG model or refining a mature systematic process, we meet you where you are and push you forward with practical tools and honest feedback.

Organised desk with financial planning documents

Map your ESG data landscape before you model it

We help teams audit their existing ESG data sources, identify coverage gaps, and design acquisition strategies that align with their specific analytical objectives rather than vendor marketing claims.

Multiple screens showing unified data dashboard

Unify fragmented ESG datasets into one coherent view

Most organisations juggle five or more ESG data providers with incompatible identifiers and conflicting metrics. We build reconciliation pipelines that resolve these conflicts and produce a single, analysable dataset.

Answer the ESG questions that keep coming up

Our clients ask us the same hard questions repeatedly: how do we handle missing data? Which materiality framework should we trust? We have distilled thousands of conversations into practical decision trees.

Deploy models that survive contact with real markets

Backtests are easy. Live implementation is hard. We help teams stress-test their ESG signals against transaction costs, liquidity constraints, and regime changes before they commit capital.

The people building your ESG analytical infrastructure

Dr. Eleanor Vance-Chen

Lead Quantitative Researcher

Eleanor spent a decade in academic finance before realising that the most interesting ESG questions were being asked by practitioners who lacked the tools to answer them. She now leads our research programme.

Recognition for advancing ESG integration methodology

Excellence in Quantitative ESG Research

2024

Innovation in Cross-Sectional Analytics

2023

Best Methodology Paper on ESG Signal Construction

2022

Outstanding Contribution to Sustainable Finance Education

2021

What guides our work every single day

Our mission

We advance the field of sustainable investing by making rigorous, quantitative ESG integration accessible to practitioners who refuse to compromise on analytical standards.

Our vision

A capital market where ESG signals are evaluated with the same empirical discipline as traditional financial factors, producing better outcomes for investors and society alike.

Intellectual rigour

We never publish a model we have not stress-tested, never recommend a data source we have not audited, and never present correlation as causation.

Radical transparency

We share our methodology, acknowledge our limitations, and welcome scrutiny. Every assumption is documented and every limitation is disclosed.

Practical relevance

Research that cannot be implemented is research that does not matter. We optimise for decisions, not citations.

Continuous improvement

We revisit our models quarterly, update our priors when evidence demands it, and treat every failed backtest as a learning opportunity.