One register. Machine-checked on every draft.
exhibit registration, labelling, and cross-referencing across affidavits, replies, schedules, briefs and court books.
A mislabelled exhibit in a filed document is not a typo. It is a credibility problem, an amendment application, and a cost order waiting to happen.
A document-heavy matter accumulates exhibits for months or years, labelled by different people, in different documents, at different stages. The failure modes are universal:
- The same label ends up meaning two different things in two different drafts.
- The same item carries different labels depending on which schedule you read.
- Cross-references rot silently as documents are renumbered and amended.
- Nobody notices until filing, or until opposing counsel notices for you.
None of this is anyone’s fault. It’s what happens when exhibit integrity depends on human memory across a long matter. Diligence doesn’t scale; the record does.
Decompose · Build · Validate · Supervise
Exhibit management reduces to atomic questions a system answers reliably: What is this item? What label does each document give it? Where is it cited? Do any two items share a label? Does any citation point to a label whose meaning is inconsistent across the file?
A master exhibit register becomes the single source of truth: every exhibit, every label applied to it, every document that cites it. On top of it sits an automated collision check: every new draft is swept for exhibit references and validated against the register before it reaches the review pile.
The register is tested against the source documents, never against recollection. Where two sources genuinely conflict about what a label means, the conflict itself is flagged as an issue for the solicitor to rule on, not silently resolved by the system.
The system never renames anything. It surfaces each collision with its evidence (this label means X in document A and Y in document B) and a human makes the call. Every resolution is logged, so the schedule that goes to court has a defensible history.
| Before | After |
|---|---|
| Exhibit integrity depends on one person’s memory | Single auditable register, machine-checked on every draft |
| Collisions discovered at filing, or in court | Collisions surfaced at drafting, with evidence attached |
| Hours of manual cross-checking per draft | Minutes, with a logged result |
| “I think that’s the right number” | “Here is the register entry, its sources, and its citation history” |
Zero legal judgment is automated here. The system never decides what an exhibit proves or whether to rely on it. It automates the bookkeeping that consumes senior time and creates filing risk, and leaves an audit trail behind it. The same pattern extends directly to discovery lists, tender bundles, briefs to counsel, and court books.