Proof

How teams launched in days, not quarters.

The honest version — what they set up on day 1, what broke, what their AI gets right today. No names are invented; placeholders are clearly marked.

Reference deployment

How our insurance reference tenant serves 1,200 FAQs at sub-second latency

A walk-through of the deployment we use ourselves to validate every release.

1,200+

indexed FAQs

across finance, insurance, healthcare — retrieved via hybrid RAG in <350 ms median

Challenge

Most insurance sites ask a visitor to compare 4–6 plans across 15 variables: sum insured, room rent, co-pay, sub-limits, waiting periods, OPD. The typical outcome: 3 tabs open, decision fatigue, and an unattended contact form.

The reference tenant needed to handle those comparison questions without sacrificing the compliance language the regulator requires.

Setup

  • Uploaded policy PDFs via /knowledge/import-document — auto-chunked into ~400-word KB entries and indexed into Qdrant.
  • Configured the Insurance tone preset plus a custom snippet: "Use Indian English. Always quote premiums in ₹. Never promise coverage — say 'subject to policy terms'."
  • Enabled the orchestrator's rules engine to hand off to a human advisor whenever the retrieval confidence fell below 0.55.
  • Turned on the embedded widget on the Hanvitt marketing site plus a staged WhatsApp Business flow for inbound leads.

Result

<1s

First answer in

84%

KB hit-rate on real visitor questions

32% of started convos

Lead captures from chat sessions

100%

Advisor handoffs (vs. hallucinated answers)

Every Hanvitt release ships only after the reference tenant passes a 50-query regression. It's how we catch regressions before you do.

Healthcare · coming soon

Replacing reception phone triage at a multi-specialty clinic (pilot in progress)

Named case study coming after the pilot wraps in Q2 2026. Below is the setup and what we're measuring.

Pilot

in progress

we'll publish full numbers once the pilot is signed off — the structure below is live today

Challenge

A multi-specialty clinic fielding 400+ phone calls/day for appointment booking, doctor availability, and insurance empanelment questions. Reception dropped ~18% of calls during peak hours.

"We don't want a chatbot that pretends to be a doctor. We want one that handles the boring 80% so reception can focus on the hard cases."

Setup

  • Enabled the Healthcare domain pack — includes the built-in disclaimer "This is not medical advice. Call 108 for emergencies."
  • Imported the doctor roster + availability templates from a CSV via /knowledge/import-csv.
  • Wired the Doctors module so visitors can check slots and book appointments without leaving the chat.
  • Configured the Concise tone preset (no filler, no emojis) because the audience is mostly elderly patients and their attendants.

Result

40–60%

Target: reception calls deflected

-15%

Target: appointment-to-visit no-show reduction

90 days

Measurement window

We'll update this case study with named metrics once the pilot concludes. No fabricated numbers here in the meantime.

Want to be the next case study?

Start free, and we'll check in at week 6 to see what you'd like us to publish (anonymised if you want).

Your partner in Growth — For Individuals & Businesses

Hanvitt Consulting & Solutions — four disciplines, one partner. AI consulting, lead generation, modern websites, and legacy modernization, shipped end to end. We also run two platforms on the side: Hanvitt.in for individuals and Hanvitt AI Platform for SMEs.

Services

Platforms

Company

Resources

Legal

© 2026 Hanvitt Consulting & Solutions. All rights reserved.

Made for businesses that never want to miss a customer.