Key takeaways
- iManage and NetDocuments are document management systems — built to store, version, and secure documents, not to read them for substance.
- A conflict check today still relies on a name-matching search and a human reviewing what comes back, even though the actual matter documents already sit in the DMS.
- An n8n + Claude workflow reads the substance of new-matter intake against the firm's existing matter documents, surfacing genuine conflicts a name search alone would miss.
- This doesn't replace the conflicts process or the partner sign-off — it replaces the manual reading time between a name-match hit and knowing whether it's actually a conflict.
- Firms running this catch conflicts earlier, with less manual document review per intake, without changing who makes the final call.
A Document Management System Is Not a Reading System
iManage and NetDocuments are excellent at what they're built for: storing, versioning, and securing every document a law firm produces. Ask either one to read those documents and tell you something about their content, and the honest answer is that they don't — they're a filing cabinet with very good locks, not a research assistant. That distinction matters most in exactly the place firms rely on it least: conflict checking.
What a Conflict Check Actually Is Today
Most firms run conflicts as a name-matching search against a database, and then a person — often an associate or a conflicts specialist — reviews whatever comes back to decide if it's a real conflict or a coincidence of names. The problem is that a real conflict isn't always a name match. It can be a related entity, a prior adverse party mentioned inside a document rather than in a case caption, or a connection that only shows up if someone actually reads the substance of prior matters. That substance already exists — it's sitting in iManage or NetDocuments right now — but nobody's reading all of it for every new intake, because that would take a person days.
The Workflow: Claude Reads What the DMS Only Stores
We build this as an n8n workflow that runs at new-matter intake: it pulls the relevant details of the incoming matter, searches and reads across the firm's existing matter documents in iManage or NetDocuments — not just names, but actual substance — and surfaces anything that looks like a genuine conflict, with the specific document and passage that triggered the flag. A person still makes the final conflict determination. The workflow's job is narrower and more honest than "clearing conflicts": it's making sure the person deciding actually saw everything relevant, instead of whatever a name-match search happened to surface.
What This Doesn't Change
The conflicts process, the ethical wall procedures, and the partner sign-off all stay exactly where they are. What changes is the amount of manual document review standing between a name-match search returning something and a qualified person actually seeing the relevant substance. That review used to take however long a human needed to search a DMS by hand. Now it takes as long as it takes an agent to read.
Frequently Asked Questions
Does this replace our conflicts counsel or process?
No. It surfaces information for a human to review and decide on — it does not clear conflicts or make ethical determinations.
Is privileged document content safe to run through this workflow?
The workflow is built to the same confidentiality standard as any system touching privileged material — documents are read and processed within a controlled environment, not exposed to a public model with data retention you don't control.
Does this replace iManage or NetDocuments?
No — both remain the system of record for every document. This workflow reads what's already stored there; it doesn't change how or where documents are managed.
How does this handle a genuinely ambiguous conflict?
It's flagged for a human, with the supporting passage attached, rather than resolved automatically — ambiguous conflicts are exactly the cases that need a lawyer's judgment, not an agent's guess.
