Content pipeline
Content pipeline running on autopilot
Keyword research and publishing pipeline for an in-house SEO stack. Indexed pages generated, linked, and maintained without a writer on payroll.
Publishing without an in-house writer
Document automation
Fee earners open contracts pre-marked, not blank
Inbound contracts and client email hit a triage layer before they hit the fee earner. Parties, dates, payment terms, liability caps, and deviations from the firm's playbook arrive already pulled out and flagged.
Client
UK mid-market law firm
Build time
8 weeks to production
Headline outcome
First-pass without a junior
The problem
Mid-market firms lose the first hour on every contract rediscovering the same structural data: who signed, what the term is, where the indemnity lives. Associates do this work because partners shouldn't, and it still costs 45-60 minutes per document. Meanwhile partner inboxes drown in 300+ emails/day with no consistent triage.
The system
A matter-aware intake layer. Contracts flow through a multi-pass extraction: a first pass pulls structural data (parties, dates, governing law, caps), a second pass scores clauses against the firm's playbook and flags deviations. Inbox triage tags every email by matter, urgency, and type and drops non-urgent items into a digest.
How it runs
The workflow, end to end.
01
Inbox classification
Every inbound email is classified by matter, urgency, and type (contract, query, status update, admin). Non-urgent items flow to a 4pm digest.
02
Contract extraction
Structured extraction pulls parties, effective date, term, governing law, payment terms, liability caps, termination, IP ownership. Outputs to a structured template the fee earner sees.
03
Playbook diff
Each clause is compared to the firm's negotiation playbook. Deviations are scored red / amber / green with the exact playbook clause referenced.
04
Redlines drafted
For amber/red deviations, a suggested redline is drafted in the firm's preferred wording. Fee earner accepts, edits, or rejects.
05
Matter write-back
Extracted structural data writes to the PMS (Clio, ActionStep) so the matter record reflects the document, not just a filename.
Outcomes
What it moves.
60min → 15min
Contract first-pass time
-70%
Inbox triage time / day
+40%
Playbook compliance rate
+25%
Matters per fee earner
Stack
What it's built on.
- Claude (extraction + playbook diff)
- Microsoft Graph API (inbox)
- Clio / ActionStep API
- Postgres
- Python
LegalContract reviewDocument automation
Mid-market firm doing 50+ contracts/month? Let's scope the first pass.
20 minutes. No slides. One process, one prototype.
Book a 20-min introMore builds
View all