We confess: three years ago, we spent more time arguing about vendor scores than building models that worked. Today, our small team helps institutional analytical teams embed environmental, social, and governance signals directly into quantitative frameworks — with documented methodology, honest limitations, and code that runs in production.
We model relative positioning within peer groups rather than absolute thresholds, capturing the variation that drives cross-sectional return differences.
Research outputs arrive as version-controlled code with reproducible pipelines, not slide decks that decay the moment they are presented.
We construct custom ESG signals from raw vendor data, test their cross-sectional predictive power, and document every modelling choice so your team can defend the methodology to any sceptic.
Learn moreFor teams with specific analytical questions, we design and execute custom research projects with defined deliverables, documented methodology, and full code handover upon completion.
Discuss scopeWorking with us means sharper questions, cleaner data, and signals your investment committee can actually trust
Documented assumptions and reproducible pipelines mean fewer surprises when a model meets live market data for the first time.
Faster iteration cycles
Production-ready code and clear diagnostic outputs let your team test variations without rebuilding infrastructure from scratch.
A structured pathway from initial conversation to independent operation, with clear milestones and no hidden phases
We begin with a structured conversation about your current ESG data landscape, analytical infrastructure, and the specific questions your team is struggling to answer. This phase produces a written assessment with honest recommendations about whether our approach fits your needs.
All pipelines, functions, and tests are documented and transferred to your team. We conduct knowledge-transfer sessions until your analysts can run, modify, and extend the models independently.
We primarily deliver in Python and R, with version-controlled repositories and documented dependencies. If your infrastructure requires other languages, we can discuss adaptation during the diagnostic phase.
Schedule a diagnostic assessment with our research team. We will honestly evaluate whether our approach fits your infrastructure, your questions, and your investment process.
Evidence-first culture
Every signal is tested against live data, every assumption is documented, and every limitation is disclosed before a recommendation leaves our desk.