Skip to content
Meridian Health · AI · 2026

A grounded assistant that drafts replies from verified policy documents — and hands off the moment it is unsure.
  • -68%First response time
  • 94%Draft acceptance rate
  • 0Unsourced claims in production

The challenge

Meridian's support team answered the same two hundred policy questions every week, each requiring a careful read of documents that changed monthly. Response times drifted past a day, and in a regulated setting a confidently wrong answer was a far worse outcome than a slow one.

What we did

  1. 01

    Built retrieval over the verified policy corpus so every draft is grounded in a specific, current, citable passage.

  2. 02

    Kept a human in the loop by design — the assistant drafts, an agent approves. It never sends unattended.

  3. 03

    Defined explicit refusal behaviour: below a confidence threshold, or outside the corpus, it escalates instead of guessing.

  4. 04

    Built an evaluation set of six hundred real historical tickets and measured accuracy against it before anything reached production.

The outcome

First response time dropped 68% while accuracy went up, because every answer now cites current policy. Agents accept 94% of drafts with light edits, and no unsourced claim has reached a customer.

They pushed back hard on letting it reply unattended. In our sector that judgement was worth more than the feature would have been.

Sarah WhitfieldVP Customer Operations, Meridian Health