REF / AUTOMATION

London Boutique Law Firm - Contract Review and Intake Automation

Built AI-powered contract review and client intake automation for a 9-lawyer London firm - recovered 22 billable hours per lawyer per week.

RoleAutomation Lead
Year2025
Outcome+22 billable hours/lawyer/week
DomainAutomation
00
STACK

Tech used.

Claude Opus 4.6Microsoft Graphn8nNetDocumentsDocuSignStripe

The Problem

A 9-lawyer commercial firm in central London was watching its margins erode under the weight of non-billable work. Lawyers were spending 8-12 hours a week each on tasks that didn't bill: reviewing routine NDAs, summarising lengthy contracts for partners, doing first-pass intake calls with new prospects, and chasing missing documents from clients.

The managing partner had been told by a Big Law cousin that AI would "take their jobs." The reality, after running the numbers, was the opposite. AI could give them back their highest-margin hours.

What I Built

A precision-targeted automation suite focused on the non-billable work, not the legal advice itself:

1. Contract review assistant. A lawyer drops a contract into a OneDrive folder; within 90 seconds they receive:

  • A markup highlighting clauses that deviate from the firm's standard terms
  • A risk summary (top 5 concerns, ranked)
  • Suggested redlines
  • Comparison to similar past deals from the firm's matter database

The lawyer reviews, accepts/edits, and sends. What used to be 90 minutes of first-pass review becomes 15 minutes of senior judgement.

2. Intake automation. Prospective clients book via a smart form that, before the call, sends:

  • A targeted intake questionnaire (different per matter type)
  • Document upload links
  • An automated pre-call summary delivered to the partner

By the time the partner calls, they have the conflict check done, key facts on screen, and a draft engagement letter ready to send. Intake calls dropped from 45 minutes to 18.

3. Document chase workflow. When a matter is missing required documents, the system auto-emails the client at days 3, 7, and 14 with a personalised reminder and a secure upload link. Status reports go to the partner; no human follows up unless the automation fails.

4. Time capture suggestions. A daily AI assistant reviews each lawyer's calendar + email + document edits and suggests time entries. The lawyer approves or edits. no more reconstructing the week on a Friday.

Contract Review Workflowfirst-pass in 90 seconds
  1. 01
    TriggerContract dropped in OneDrive

    NDA, MSA, or commercial agreement lands in the matter folder.

  2. 02
    StepClaude markup generated

    Clauses deviating from firm standard are highlighted; risks ranked top-5.

    ~90 seconds
  3. 03
    StepSuggested redlines drafted

    Track-changes proposals plus comparison to similar past deals from the firm's matter database.

  4. 04
    DecisionLawyer reviews & exercises judgement

    Senior lawyer accepts, edits, or escalates. AI never sends anything client-facing.

    90 min → 15 min
  5. 05
    OutputFinal draft sent to client

    Engagement letter, redlines, or signed copy via DocuSign. File note auto-written for matter record.

+22h
Billable hours/lawyer/week
−83%
First-pass NDA review time
90m → 15m
60%
Time capture from AI suggestions
£480K
Annualised billable lift
9 lawyers
Per-Lawyer Weekly Time
MetricBeforeAfterΔ
Billable hours28h50h+22h
Routine contract review8h1.5h−6.5h
Intake call duration45m18m−60%
Time-capture admin3h30m−83%
Document chasing2h0−2h

Outcome

The firm added roughly £480K of annualised billable capacity without hiring. The managing partner used the savings to fund a junior associate's training programme and to drop one of two paralegals. who left with a referral to a friendly firm and a glowing reference. Lawyers are happier doing legal work instead of admin; client satisfaction scores are up because turnaround time on routine matters dropped.