How Meridian Consulting Cut Email Response Time by 80% with AI Labeling
Challenge
Meridian Consulting — a 12-person management consulting firm specializing in operational efficiency for mid-market companies — had an inbox problem that was quietly killing their responsiveness.
The firm managed relationships across 40+ active clients at any given time. Every day, hundreds of emails arrived across three shared inboxes: a general inquiries address, a client services address, and a proposals inbox. The volume was manageable — but the chaos wasn't.
Emails from prospective clients sat unread for hours because they were buried under vendor newsletters, LinkedIn notifications, and automated billing alerts. Urgent requests from existing clients looked identical in the inbox to routine status updates. Junior consultants spent 90 minutes every morning just triaging — reading subject lines, deciding what mattered, forwarding to the right person.
The firm's managing partner, James Okafor, put it plainly: "We were spending more time managing our inbox than managing our clients. And the scary part was we didn't even know what we were missing."
Two client inquiries had gone unanswered for over 48 hours in Q4 2025. One converted elsewhere. The other complained.
It was time for a change.
Approach
Meridian connected their three Gmail inboxes to InboxPilot and spent one afternoon setting up their labeling system. Rather than using the default labels out of the box, they mapped their labeling structure to how their business actually worked.
They created 11 custom labels:
- New inquiry — first-time contact from a potential client
- Existing client — correspondence from current engagements
- Urgent — anything flagged as time-sensitive or requiring same-day response
- Proposal request — requests for scope, pricing, or proposals
- Billing — invoices, payment confirmations, expense approvals
- Contract — agreements, amendments, legal documents
- Scheduling — meeting requests, calendar invites, rescheduling
- Marketing — newsletters, promotions, vendor outreach
- Applicant — job applications and recruiting correspondence
- Internal — team communications routed through shared inboxes
- Misc — anything that didn't fit another category
Each label had a plain-English description. "New inquiry" was defined as: "First contact from someone who has never worked with us before, asking about our services, pricing, or availability." The AI used those descriptions — not keywords — to make labeling decisions.
They enabled InboxPilot's inbox scanning feature so the AI could learn from their existing email history, improving accuracy from day one. Within 48 hours, the labeling accuracy across all three inboxes was consistently high enough that the team stopped double-checking.
Week two, they layered in email actions on top of the labels. Marketing emails were archived automatically. Billing emails were forwarded to their operations coordinator. Scheduling emails triggered a response with their Calendly link. New inquiry emails were starred and flagged for review within the hour.
Solution
Within four weeks, Meridian had fully replaced their manual triage process with an automated labeling and routing system across all three inboxes.
What changed:
- Every incoming email was labeled before anyone opened their inbox
- Urgent and new inquiry emails surfaced immediately regardless of when they arrived
- Marketing, billing, and scheduling emails were handled automatically without human review
- Junior consultants were freed from morning triage entirely
- The whole team worked from a labeled, prioritized inbox instead of a flat chronological list
- No email required someone to read it just to decide who should read it
The operations coordinator, who had been spending significant time on email routing, was reassigned to client-facing work within the first month.
Results
What’s next
With the labeling system running smoothly, Meridian is now expanding their InboxPilot setup to include:
- AI drafting for new inquiry responses, trained on their proposal templates and service descriptions
- Auto-responding to scheduling requests using their standard availability windows
- A custom label for "at-risk client" — emails showing signs of dissatisfaction — flagged for immediate partner review
Their next milestone: full AI drafting across all three inboxes by Q3 2026, with the goal of reducing email-related work to under 20 minutes per day per consultant.
This case study is based on a composite of real customer experiences using InboxPilot’sAI email automation features.
You have been doing it manually for too long
Connect your Gmail or Outlook inbox in one click, train InboxPilot on your business data, and start receiving replies that sound exactly like you – ready to send or sent automatically.
