Document intelligence
A published PDF is not an index. Institrace is the layer that turns the public record into a single searchable corpus, then keeps that corpus current.
We only index public bodies. A supplier, developer, or care-home Ltd appears when a public file or a published identifier attaches it. We do not run a private-company directory.
MethodologyHow records attachWorked example: Birmingham CouncilFAQ
How it works
Public bodies in the UK are required to publish committee minutes, agendas, registers of interests, spending records, decision notices, accounts and FOI responses. They do. The papers sit on ModernGov sites, board portals, disclosure logs and file stores that do not talk to each other. Many of the files are image PDFs. A search box on one council website cannot see another council, an NHS trust, a BID or a company filing.
Institrace collects those papers on a schedule. Each document is stored with its source, type, original URL and a hash taken at the moment of indexing. Image PDFs are passed through OCR so the text exists as text. Full-text search then runs across the whole corpus at once: currently 1,039,997 documents, 900,134 of them with searchable text. The original file stays at the source. Every hit links back to it.
What it does
It makes a query that no official portal can answer. A name, a contract, a street, a committee, searched once, across authorities and institution types. It also makes change visible. If a paper is replaced or removed, the hash from the first index remains. The record of the record is itself a record.
This is not a clipping service and it is not a copy of WhatDoTheyKnow. It is an institutional index: papers classified, dated, attached to a public body, and held next to the people, land and offshore layers that use the same graph. No other UK service runs that stack on the same records.