Evidence-grounded discovery for accountability work

Surface evidence-grounded signals quickly, carefully, and at scale.

Accounter reads large document sets in minutes and produces a permanent, structured record of evidence-linked signals that warrant attention.

Multiple ways to explore. One standard of evidence.

Built for accountability work: journalism, oversight, investigations, research.

Request access

What you get

A ledger of signals, each linked to verbatim evidence. Only surfaces signals whose excerpts are validated against extracted text. Multiple export options. Designed for repeat runs and longitudinal comparison.

Sample Signals ledger

Mar 2022

Revised cost estimate 40% above approved figure

"The revised estimate for the Northern Rail Link is now £4.2bn, representing a 40% increase on the figure approved by Parliament in 2019."

Select Committee Evidence
Page 7
Northern Rail Link

Nov 2023

Final cost projection exceeds £5bn; delivery delayed to 2028

"The Department now expects final costs for the Northern Rail Link to exceed £5.1bn. The opening date has been revised from 2025 to 2028 at the earliest."

NAO Report
Page 12
Northern Rail Link

Jul 2024

PAC warned of optimism bias in original business case

"The Committee remains concerned that the Northern Rail Link business case did not adequately account for cost escalation risk. We note that internal modelling was not disclosed to Parliament at the time of approval."

Public Accounts Committee Report
Page 4
Northern Rail Link

Ways to discover

Discover more from different angles. Accounter lets you explore large document sets in multiple ways — without changing methodology or losing traceability.

Explore broadly

Surface risks, tensions, and deviations across entire document sets — even when they're buried or euphemistically phrased.

Focus on a topic

Examine how a specific issue (e.g. finance, safety, governance) appears across documents, committees, or time.

Follow a thread

Take a handful of notable excerpts and ask: "Where else does this kind of thing appear?"

Check explicit references

Find every direct mention of a name, project, or phrase — with page-accurate citations.

Rediscover at any point

Run a fresh discovery, narrow the scope, or follow a new line of inquiry — using the same source material and the same evidence rules.

The approach

Increase the chance of surfacing important signals. Keep human judgment explicit.

Processes full documents

Body text, appendices, footnotes, extracted tables. No content weighting — appendices and footnotes receive the same scrutiny as main sections.

Biased toward coverage

Designed not to miss signals. You decide what matters; the system ensures you see the options.

Isolated workspaces

Each account has a private workspace. Documents and outputs are not shared between customers.

Why this complements manual review

Different failure modes. Use both.

Reads exhaustively

Footnotes, appendices, extracted tables
No fatigue, no time pressure
Doesn't skip what looks routine

Surfaces what you didn't search for

Keyword search won't find what you don't know
Surfaces related issues even across documents
No need to guess the right terms upfront

Preserves the evidence

Every signal linked to verbatim text
No inference, no judgment, no hallucination
Structure and validation that raw AI lacks

Fast and repeatable

Processes years of documents in minutes
Consistent, structured outputs across runs
Built to support longitudinal comparison

Constraints

What this does — and what it deliberately does not.

It does

Validate every excerpt against extracted text before surfacing
Link every signal to a verbatim excerpt and precise location
Behave consistently across documents and years

It does not

Infer intent, wrongdoing, or "what they really meant"
Paraphrase or alter source text
Surface signals without validated evidence

Request access

Access is currently limited. If you're working with large sets of public documents and would like to use Accounter, tell us a bit about your use case.