AI Agents

Context

Built for insurance within the Renewals & Upsell stack and used by retention, renewals, and contact-centre teams, this agent focuses on the moment a customer or intermediary sees a changed premium and asks “why?”. Its job is to explain premium changes to agents and customers with clear underlying drivers, so the new price doesn’t feel arbitrary.

What it does

The agent compares the expiring policy with the renewal (or mid-term adjustment) and breaks the premium difference down into driver-level components, aligned to your rating logic. It shows which inputs changed between the two terms—such as risk factors, cover limits and deductibles, discounts and loadings, usage/telematics signals, taxes or fees—and how each of those contributed to the final increase or decrease. From that decomposition, it generates simple, channel-ready explanations: short snippets for letters, emails, and portals, and richer talking points for agents and call-centre scripts, so everyone is working from the same, transparent “why it changed” story.

Core AI functions

The core capability is pricing explanation. The agent reads structured outputs from the rating engine and policy system for both old and new terms, identifies which factors moved, and calculates their contribution to the premium delta. On top of that, it applies controlled natural-language generation, mapping each driver into pre-approved explanation patterns and phrasings, so the result is consistent, compliant wording rather than free-form text.

Problem solved

Today, many renewals arrive as a new premium with no clear rationale, forcing agents and contact-centre staff to reverse-engineer rating screens under pressure or resort to generic lines about “market conditions” and “inflation”. Customers experience changes as opaque and unfair, which drives complaints, shopping around, and avoidable cancellations—even when the underlying reasons are legitimate and already encoded in the rating logic.

Business impact

By explaining driver-level changes transparently, the agent reduces complaints and cancellations linked to “unexplained premium increases” and makes renewal conversations more constructive. Customers and intermediaries get a clear view of what changed and why, which supports trust and opens the door to adjusting cover, deductibles, or optional benefits instead of lapsing outright. That improves persistency, protects premium income, and helps demonstrate fair-value, transparent pricing to regulators and consumer bodies.

Integration and adjacent use cases

Integration complexity is low: the agent needs read access to rating-engine outputs and policy data for expiring and renewal terms, and a way to surface its explanations into policy admin, CRM, portals, and communication templates.

Common combinations in this stack:

• Lapse Risk Predictor to prioritise which renewals most need strong explanations and targeted retention treatment;

• Usage-Based Data Interpreter (Telematics) to translate telematics or IoT usage patterns into understandable behavioural drivers within the premium explanation;

• Product Suitability Re-Matcher to propose better-fitting products when the explanation reveals mismatches between current cover and customer needs; and

• Cross-Sell / Upsell Trigger Agent to suggest relevant add-ons or alternative covers once the premium change has been clearly explained, so renewal and growth conversations are grounded in a transparent “here’s what changed and what you can do about it” narrative.