Skip to content
Email Automation · Legal

Giving a litigation team back 31 hours a week

Paralegals were spending half their day sorting mail and re-typing intake forms. We automated the triage, the drafting and the filing.

Client
A 12-attorney litigation firm
Industry
Legal
Timeline
4 weeks to production

Results

31h

Reclaimed per week

Across four paralegals, measured by time-tracking before and after.

100%

Intake logged same day

Previously 61% within 24 hours, with a 3-day tail.

0

Missed deadline notices

Court-deadline emails now auto-flagged and calendared on arrival.

The problem

What we found before anything was built.

  • 01

    A shared intake inbox received 300+ emails a day: new client enquiries, opposing counsel correspondence, court notices and vendor noise, all interleaved.

  • 02

    Paralegals manually re-keyed intake details from unstructured email bodies and PDF attachments into the practice-management system.

  • 03

    Court deadline notices occasionally sat unread for a day — an unacceptable risk that was being managed by vigilance alone.

  • 04

    Standard acknowledgement letters were written by hand, one at a time, from a Word template.

The solution

What we built, and why it was built that way.

  • 01

    Classified every inbound message into one of nine categories with a confidence score, routing each to the right queue and owner.

  • 02

    Extracted structured intake fields from email bodies and PDF attachments — parties, matter type, jurisdiction, key dates — into a validated schema before any write occurred.

  • 03

    Auto-created the matter record in Clio with documents filed to the correct folder and named to the firm’s convention.

  • 04

    Detected court-deadline language, calendared the date with statutory lead-time reminders, and escalated to the responsible attorney in Slack immediately.

  • 05

    Generated acknowledgement and engagement letters as drafts for one-click attorney review. Nothing sends to a client without a human clicking send.

Architecture

How a single request moves through the system

Microsoft Graph APIOpenAI (document extraction)Clio APIn8n orchestrationAzure Blob StorageSlack Bolt
  1. 01Email received
  2. 02Classify + score
  3. 03Extract fields
  4. 04Validate schema
  5. 05Create matter
  6. 06Draft response
  7. 07Attorney review

Every step is idempotent and retried with exponential backoff. Failures land in a dead-letter queue with a Slack alert rather than disappearing, and the confidence gate routes anything uncertain to a human with full context attached.

The part I did not expect was the audit trail. Every intake is timestamped, every extraction is reviewable, and I can show a client exactly when we received something.

Managing Partner·Litigation firm · 12 attorneys
30 minutes · No obligation

Find out what your team could stop doing by next month.

Bring one process that frustrates you. We’ll tell you whether it can be automated, roughly what it costs, and what it would give back.

  • We map where the hours go

    A quick walk through your day-to-day to find the processes eating the most capacity.

  • We rank by return

    You get an honest read on what’s worth automating first — and what isn’t worth it at all.

  • You leave with a plan

    A written recommendation and a rough number, whether or not you work with us.