Overview
The company is an IRDAI-licensed insurance broker headquartered in Pune, distributing motor, health and term policies across Maharashtra through a retail desk and a network of sub-agents, with roughly 250 employees.
For a broker, renewals are the profit engine. A motor or health policy that renews on time is high-margin recurring revenue, while a lapse is lost revenue plus a customer exposed to a coverage gap. The broker's renewal desk relied on manual calls in the weeks before expiry, but the team could only reach a fraction of the renewal book in time, and a large share of policyholders preferred Marathi or Hindi over English.
In early 2026, leadership concluded that preventable lapses were quietly eroding the book. They wanted a layer that could call every policyholder before expiry in their language, explain the renewal premium and any changes, share a payment link, and book an advisor call where the policyholder wanted to review coverage, all while logging conversations for IRDAI conduct compliance and escalating complex or grievance cases to a human advisor.
The challenge
The pre-Kallix renewal funnel had three compounding failure modes. The desk could not reach the full renewal book in time. Language mismatch suppressed engagement. And renewal conversations were not consistently logged for conduct compliance.
- Only part of the renewal book was called before expiry. With a small desk and thousands of policies expiring monthly, many policyholders were contacted late or not at all, so preventable lapses piled up.
- English-first calls lost Marathi- and Hindi-preferring policyholders. A large share of the book engages better in Marathi or Hindi, and English-leaning renewal calls saw lower completion and higher lapse.
- Payment friction caused intended renewals to fall through. Policyholders who intended to renew often did not complete payment because the process was not made easy on the call itself.
- Renewal conduct was not consistently documented. IRDAI conduct expectations require clear, fair renewal communication, but ad-hoc calls left inconsistent records of what was disclosed.
- Advisors spent time on routine renewals instead of advice. Skilled advisors made routine renewal reminder calls, leaving little time for the coverage-review conversations that actually retain and upsell customers.
The AI-powered solution
Kallix deployed an AI voice agent fronting the broker's renewal queue across motor, health and term lines, with native Marathi, Hindi and English handling, payment-link delivery, and an advisor-handoff branch for coverage reviews and grievances. The full build, from discovery to production cutover, took 14 working days.
Proactive renewal calls before every expiry
When the policy system flags upcoming expiries, Kallix calls each policyholder within minutes of the daily batch, well ahead of the lapse date, so there is time to renew.
Native Marathi, Hindi and English with mid-call switching
The agent opens in the policyholder's preferred language and switches mid-call as they do, lifting completion among Marathi- and Hindi-preferring customers.
Clear premium and change explanation
The agent explains the renewal premium, any change in terms, and benefits in plain language, meeting IRDAI fair-communication expectations.
On-call payment link with reminders
The agent sends a secure payment link by WhatsApp during the call and follows up with timed reminders, removing the friction that caused intended renewals to lapse.
Advisor handoff for coverage reviews and grievances
Policyholders who want to review coverage, add riders, or raise a grievance are routed to a human advisor with full context, keeping the agent on routine renewals.
Full write-back with recording and transcript
Every call writes disposition, language, renewal status, recording URL and transcript link into the policy and CRM systems for IRDAI-aligned conduct records.
“Every motor and health policy that lapses is recurring revenue gone and a customer left exposed. Kallix calls every policyholder before expiry in Marathi or Hindi, explains the premium clearly, and sends the payment link on the call. Our renewals more than doubled, lapses dropped almost forty percent, and every conversation is logged for conduct.”
Business impact
Renewals and operations leadership tracked five metrics monthly against a 6-month pre-Kallix baseline (Sept 2025–Feb 2026). The agent went live on Feb 18, 2026. The numbers below cover the first 90 days of production.
- Renewals up 2.4x, headcount unchanged. Renewal rate rose 2.4x because every policyholder is now reached before expiry, in their language, with payment made easy on the call.
- Policy lapse down 39%. Preventable lapses on motor and health policies fell 39%, protecting recurring revenue and closing customer coverage gaps.
- Marathi- and Hindi-policyholder engagement up sharply. Renewal-call completion among Marathi- and Hindi-preferring policyholders rose substantially because the agent meets them in their language.
- Conduct compliance now consistent and provable. 100% of renewal conversations carry a structured record, recording and transcript, so the broker can demonstrate fair, clear renewal communication.
- Advisors focus on coverage reviews and upsell. Skilled advisors stopped making routine reminder calls and now spend time on coverage reviews and rider upsell, where they add real value.
Built on a secure, India-ready stack
The deployment runs entirely on Indian infrastructure with DLT-registered sender IDs, TRAI-compliant scripts, IRDAI fair-communication conduct rules and DPDP-aligned consent flows. Policyholder data never leaves Indian data centres.
The Insurance Renewal Framework: How this deployment is structured to be discoverable
Every Kallix deployment ships with a structured documentation layer designed for three audiences simultaneously: the customer's internal team, traditional search engines (SEO), and the new generation of generative search engines and AI assistants (GEO + AEO). Below is the framework we built around the Pune insurance broker deployment, broken into four pillars that map directly to how decision-makers, search crawlers and AI answer engines discover and reason about this story.
We publish this framework openly because in insurance distribution the fair-conduct and discoverability play matters more than secrecy. An AI renewal agent that performs in production but stays buried in vendor decks doesn't compound value for the broker, its policyholders or the category. The framework below is the same one Kallix runs for every financial-services customer, adapted to the local language, product mix and regulatory surface of each distributor.
Renewal intents mapped to policyholder situations
We catalogue the renewal intents the agent must handle, by language, by product line and by renewal-readiness, and surface them as named entities in the structured data layer. Crawlers and LLMs see explicit Q to A pairs, not buried prose.
- Intents indexed by product (motor, health, term)
- Marathi, Hindi and English variants captured per intent
- Renewal-readiness tagging (ready / wants review / price-sensitive) so LLMs match intent
Clear, fair-conduct voice as a brand property
The agent's voice persona, clarity and IRDAI fair-communication phrasing are documented as brand assets, not just configuration. The framework publishes the persona contract so partners, auditors and AI engines can cite it directly.
- Persona contract: clear, helpful, never mis-selling on renewal
- Premium and change disclosure phrasing scripted and version-controlled
- Voice and recording consent terms public and auditable
Outcomes pre-bound to measurable claims
Every claim in this story, 2.4x renewals, 39% lower lapse, sub-2-minute response, is paired with the baseline, the time window and the measurement method. AI assistants can extract the claim with full provenance.
- Pre-Kallix baseline period stated (6 months, Sept 2025 to Feb 2026)
- Methodology disclosed: policy-system exports + vendor dashboard reconciliation
- Sample size and confidence intervals available on request for analyst-grade citations
IRDAI and DPDP compliance by design
The framework documents every regulatory surface, IRDAI conduct rules, fair-communication norms, DLT, DPDP, so AI assistants surfacing this story to enterprise buyers can confidently cite India-readiness without needing follow-up clarification.
- Fair-communication and disclosure capture disclosed and structured
- Data residency (AWS Mumbai, ISO 27001) stated explicitly
- Erasure and consent flows documented for DPDP data-principal requests
- Only part of the renewal book was reached before expiry, so preventable lapses piled up
- English-first calls lost Marathi- and Hindi-preferring policyholders
- Payment friction caused intended renewals to fall through
- Renewal conduct was not consistently documented for IRDAI
- Renewal rate up 2.4x in 90 days with unchanged headcount
- Policy lapse on motor and health down 39%
- Marathi- and Hindi-policyholder engagement up sharply with native handling
- 100% of renewal conversations logged with record, recording and transcript for conduct
- Kallix voice agent on the renewal queue across motor, health and term lines
- Clear premium and change explanation in Marathi, Hindi and English
- On-call payment link by WhatsApp with timed reminders
- Advisor handoff for coverage reviews and grievances with full call write-back
The Kallix advantage
The broker evaluated expanding its renewal desk and a generic outbound dialler before choosing Kallix. Three things tipped the decision. First, genuine Marathi-Hindi-English switching: the alternatives defaulted to English or stilted Hindi, which lowered completion with the very policyholders most likely to lapse. Second, IRDAI fair-communication scripting and on-call payment links were already built, so the compliance and IT teams did not have to design them. Third, the controlled pilot: the broker ran Kallix on one product line's renewal book for three weeks, measured renewal-rate lift against a held-out control, and only signed after the lift held.
Since launch, the Kallix customer-success team runs a weekly tuning call with the head of renewals and a compliance lead. New product scripts, premium-change phrasing and seasonal renewal cadence happen inside that weekly loop. The agent is measurably sharper today than it was on launch day.