We do not publish your view. We build the layer underneath it — documents turned into structured data, sector intelligence assembled from primary filings, and forensic and quantitative models built to your brief. Your analyst writes the note. Your name goes on it.
Everything below runs on the same extraction and modelling stack that produces the public terminals on this site. What you see live is the proof — not a mock-up of it.
Five to ten years of a company's own filings, normalised. Group map, related-party trail, accounting-quality flags, every claim linked to its source document.
Annual reports, DRHPs, concall transcripts, offer documents and filings turned into clean, typed, queryable records — delivered to your schema.
Industry structure, capacity, pricing, margin architecture and entry economics — assembled from primary filings rather than recycled from secondary sources.
Prospectus forensics at scale. Claim extraction, related-party and promoter-conduct checks, OFS context, lock-in mapping and DRHP-to-RHP delta.
Three-statement, valuation, debt and project finance, transaction models and the forensic battery — built to your brief and wired to your data rather than ours.
Ongoing delivery of extracted, normalised financial and disclosure data across a coverage universe you define.
Most firms that need this capability do not want to build a data desk to get it. Private label means the extraction, the sector work and the models arrive in your template, in your format, with no attribution back to us.
You keep the client relationship, the editorial voice and the responsibility for the view. We stay invisible, and we do not compete with you for your clients.
If you have a backlog, the only question that matters is how fast it clears. These are measured figures from production work, not capacity estimates.
Ten or more filings of 500+ pages each, extracted end to end in a single four-to-five hour working session.
A single InvIT offer document of 2,400 pages taken through full extraction without splitting the run.
Seventy-plus prospectuses extracted through the engine across thirty-three sectors and roughly twelve families.
Several hundred offer documents read by hand before any of this was automated. The frameworks come from that, not from a template.
The terminals on this site are produced by the same pipeline and updated continuously. The proof is the running system.
Tolerant parsing designed against eighteen classes of malformed output, because real filings break naive pipelines.
Engagements are scoped individually. Send the shape of the problem — the documents, the universe, the format you need it in — and you will get a direct answer on whether we are the right fit and what it would take.
FinMinutes supplies data, extraction, industry analysis and quantitative models. These are inputs and tools. We do not issue recommendations on individual securities, and nothing delivered under these services constitutes investment advice to any person.