Investment research · US & India · Long horizon
We find where value is migrating.
Markets don't reward companies for what they are. They reward them for where they're going — up the value chain, out of commodities, into pricing power. Esploro is being built to see that migration early, and to prove it from the primary source.
Impersonal research & methodology — the same for every reader, never personalized advice. See disclosures.
The idea
Value doesn't sit still. It migrates.
A concept named by Adrian Slywotzky in 1996: profit steadily flows away from outdated business designs toward the ones the market will pay for next. We productize that idea as a live, evidence-grounded lens.
Commodities compress
A business that sells an undifferentiated product is a price-taker. Its margins are hostage to input costs, cycles, and the next low-cost entrant. The market pays little for that.
Some firms climb
A few move up the value chain — from bulk to specialty, from part to system, from input to brand. They win pricing power, stickier customers, and defensible margins.
The re-rating lags
The market usually prices the old, commodity version of the company for years after the climb has begun. That gap — between what a firm is priced as and what it is becoming — is what we look for.
The method
Everyone has the screener. We read the evidence.
Financial ratios are public and identical for everyone. The edge isn't the number — it's the revealed behaviour behind it: what management said versus what they actually did, what's been funded versus merely announced, what's protected by patents and partnerships.
So our framework weights revealed behaviour far above the baseline financials — and every observation is traced back to a dated primary source you can open yourself.
Indicative weighting of our research framework.
Auditability
Every call traced to the filing sentence
We don't summarize a company. We locate the exact sentence — in a concall transcript, an annual report, a regulatory filing — where the evidence lives, and we date it.
“A claim you can't trace to a source isn't research. It's an opinion wearing a suit.”
Reading the numbers
Structural, not cyclical.
The hardest call in value migration is telling a durable, mix-driven margin gain from a lucky turn in the commodity cycle. A structural climb holds through the down-cycle; a cyclical one gives it all back. Separating the two is the discipline that keeps us out of value traps.
Illustrative operating margin (%) — not any specific company.
Free · concept-level
The Value Migration Field Guide
A plain-English guide to the forces moving profit up the value chain — patent cliffs, trade agreements, import substitution, industrial policy — and how to read a company for the signature of a real climb. The framework, not a list of tips.
Writing
Insights
Essays on the method and the idea: how to tell a commodity from a specialty, how to separate a structural margin gain from a lucky cycle, and how to read management's say/do gap.
One lens, every market
Value migration isn't a country. It's a pattern.
The framework is market-agnostic — give it a universe of listed companies and their filings and it finds the same signature of a business climbing the value chain. We're building coverage one market at a time, starting where the data is open and a public track record can begin today.
United States
The reference equity market, open primary data (SEC EDGAR), and the place we can publish impersonal research and build a public, timestamped track record now.
India
Where the framework was built and pressure-tested. Registered research for Indian investors follows SEBI Research Analyst registration.
More markets
Next: Canada · European Union · United Kingdom. The funnel is market-agnostic — each sequenced to its own rules.
What is actually published here
Four surfaces, and what each one is for.
This site is deliberately small and deliberately layered. Rather than one long pitch, it separates the framework from the evidence and the evidence from the vocabulary, so you can enter at whatever depth the question needs and stop when you have the answer.
Framework
The method and the field guide
The method sets out the Analysis Bridge, the five-rung triangulation chain, and the four-part migration signature — how a claim becomes a read. The field guide works the other way round, starting from the external forces that push value up a chain: a patent expiring, a tariff line changing, a domestic buyer deciding to stop importing.
Evidence
The US research book
Company reads apply the framework to named US-listed businesses and show the working: a dated say/do ledger, the triangulation chain rung by rung, operating-margin and return history, and the filings each observation came from. The reads that end in a pass are published alongside the ones that do not, because the discipline only means something if the misses are visible too.
Vocabulary
Glossary and FAQ
The glossary defines each term once, in full, so an essay can use it without re-explaining it; the FAQ answers the questions people actually arrive with, including the uncomfortable ones about registration status and what this is not.
Scope
Markets, coverage and services
Markets states, market by market, what may be published and on what legal basis — including the markets we deliberately do not cover yet. Consulting is where the same engine is pointed at somebody else's universe.
Esploro Group · open source
The software we had to build to do the research.
Reading thousands of filings, backtesting a signal without fooling yourself, and letting software act on a finance system without losing the audit trail are all engineering problems before they are research problems. Esploro Group open-sources the tools it built for them, each on its own site under this domain. They are free to use, and nothing on them is investment advice either.
sigc
MITThe Quant’s Compiler
A type-safe DSL and Rust-native runtime for quantitative strategies: write a `.sig` file, type-check it at compile time, run the same binary in research and in production.
Strategies are expressed in a small declarative language with data, params, signal and portfolio blocks, executed over Polars/Arrow, and cached by content hash so an unchanged computation is never repeated. The research binary and the production daemon are the same artefact, which is what makes a backtest reproducible rather than merely repeatable.
closegate
Apache-2.0The policy gate for finance AI agents
An open-source policy gate, audit log and human-in-the-loop approval framework that sits between an AI agent and anything it can change in a finance system.
It enforces segregation of duties, materiality thresholds and sensitive-account routing at a single chokepoint, writes a tamper-evident audit log, and ships as a Python library, an MCP server and a container image. The point of an open gate is that the control you rely on is one you can read.
BondFoundry
MITAI on the buy-side fixed-income desk
An assistant for the buy-side fixed-income desk, built as a working reference implementation of the FINOS AI Governance Framework v2.0 rather than a demo of one.
Every action the assistant can take passes a pure-function policy gate that routes by reversibility, every decision is written to a database-enforced append-only audit chain, and approvals travel as signed envelopes the agent cannot forge. It sits alongside an existing order-management system rather than replacing it.
Building in the open · a 5–10 year project
Be early to the research, not the noise.
Join the waitlist for the method, the field guide, and our writing on where value is migrating. When the registered research goes live, you'll be first to know. No tips. No spam. No stock calls.
No spam, no tips, no stock calls. Unsubscribe anytime. We handle your data per our Privacy Policy and will never sell it.