Finance inside the governed matter

Financial analysis with the evidence and calculation route kept visible.

LegalRAG Pro Finance brings historical financial information, comparable-company analysis, deterministic calculations, source evidence and constrained AI-assisted review into the same governed matter as the documents, chronology, issues, tasks and professional decisions.

Request a Finance walkthrough See the wider product

LegalRAG Pro is legal technology, not an accounting or investment-advice service. Financial outputs remain subject to professional review.

64-second Finance workflow

Finance is the financial-evidence layer of the same legal matter.

Follow a material figure through source basis, calculation route, assumptions and limitations, then return the finding to the issue, investigation and supervised work product.

This R1 film uses authentic Finance workspace capture. A fuller interactive Issue → Finance → Task/Report recapture remains a future enhancement.

Finance practitioner training

How to use Finance in LegalRAG Pro.

A complete synthetic-data walkthrough of the governed Finance workspace: historical financials, comparable-company analysis, deterministic calculations, source evidence, limitations, exact traceability, exports and the constrained Finance AI boundary.

How to use Finance in LegalRAG Pro — 6 minutes 38 seconds.

Finance workflow · 1:04. Source-linked financial evidence remains part of the same legal matter.

One evidential chain

From source record to reviewable financial proposition.

Finance is not a disconnected spreadsheet or chatbot. It is bound to a governed Finance workspace and can be inspected through the same professional-control principles as the rest of LegalRAG Pro.

Source evidenceProvider, source, version, publication date and document coordinates.

Financial observationsHistorical values and governed period identities.

Deterministic calculationFormula, inputs, result identity and status.

Analysis & comparisonPeer summaries, target position and explicit limitations.

Professional reviewTraceability, frozen report projection and export.

What practitioners can inspect

A proper Finance workspace, not a single number generator.

Historical financialsReview exact stored observations by period, including revenue, profit measures, cash, staff costs, employees, technical staff and shareholders' funds where present in the governed dataset.

Comparable-company matrixInspect company-by-metric cells with status, value classification, period and source coordinates rather than flattening the comparison into an opaque score.

Peer statisticsReview mean, median, minimum and maximum only where the included peer data passes the deterministic status rules.

Target positionSee the target's relationship to the governed peer summary with the underlying target cell and summary identity retained.

Calculation traceInspect the formula, calculation version, input facts, observations and evidence bindings behind derived results.

Source & evidence registerTrace financial observations to provider/source/version details and, where available, document snapshot, page and bound-text hash.

LimitationsKeep source conflict, stale data, insufficient data and assumption-required states visible rather than silently filling them.

Exact traceabilityInspect outgoing references and mechanical backlinks between frozen Finance objects.

Reports & exportGenerate deterministic Markdown and HTML reports directly from the frozen Finance report projection.

Deterministic arithmetic

The language model is not trusted to do the maths.

Core Finance calculations are implemented as deterministic Decimal arithmetic. The current calculation layer includes revenue growth, EBITDA margin, equity value, net debt, enterprise value and multiples.

Why that matters for legal work

Repeatable

The same governed inputs and calculation version produce the same numerical result.

Inspectable

The formula and the exact input fact identities remain available for review.

Separate from AI prose

The Finance code expressly states that no LLM arithmetic is permitted. AI can work over frozen authority; it does not become the calculator.

Constrained Finance AI

Ask about the frozen financial authority without letting the model rewrite it.

The Finance AI boundary is deliberately narrow. It can select claims from the frozen comparable-company analysis and evidence manifest, but the model is instructed not to retrieve new figures, recalculate, normalise, rank, infer missing metrics or change a governed analytical status.

In particular, the Finance AI is told not to create valuations, target prices or buy/sell/hold advice. Unsupported or ambiguous questions are designed to return an unavailable result rather than invented financial certainty.

Designed for disputes and substantial records

Bring the financial question back into the case.

Where a matter turns on historic performance, loss, company position, financial comparison or a contested figure, the useful question is not only “what is the number?” but “where did it come from, what calculation produced it, what conflicts with it and what am I prepared to rely on?”

Evidence-linked

Financial observations and calculations retain their source and authority coordinates.

Status-aware

Source conflict, stale data, assumption requirements and insufficient data can block overconfident conclusions.

Matter-bound

A Finance workspace is explicitly bound to the active legal matter rather than floating outside the case context.

Professional control

Financial analysis is working material until the professional adopts it.

LegalRAG Pro can help organise and interrogate financial material, but it does not replace the solicitor, forensic accountant, expert or other professional whose judgment the matter requires.

Request a Finance walkthrough Review trust & due diligence →