AI & Machine Learning · Insurance

Aclaro — Insurance Vehicle Recommendation Engine

Aclaro's analysts change vehicle-insurance discount rules in Excel, and the Drools engine applies them in real time, with no developer and no redeploy.

Discount rules that kept changing

Aclaro evaluates customer campaigns by VIN, recommends service actions and calculates retention discounts based on product age, end-of-month timing and financing status. Those business rules changed often, and non-technical users needed to change the recommendation logic without a code deployment.

Rules in spreadsheets, Drools underneath

We built the engine on JBoss Drools (BRMS), with a JSON REST API for downstream systems. Business analysts maintain the rules in Excel spreadsheets. The engine evaluates each customer’s campaign history against the VIN selection criteria, applies the product-age, retention, end-of-month and finance discounts, and routes service actions to the right queue.

A Facts Table evaluation model handles complex rule chaining, SAM2 rule sets cover advanced scenarios, and customer retention profiles are managed in the same system.

Refined version by version

The rule sets went through versions v1.0 to v4.0, and the VIN selection rules alone have 3+ iterative versions. Analysts now change vehicle selection and discount rules without a developer, and the engine returns multi-factor recommendations in real time, so each campaign gets its own retention offer.

Building something similar?

If you are planning insurance pricing or discount rules your analysts can change themselves, book a 30-minute call. We will walk you through how we built this one, what it would take for yours, and who would work on it.

Book a 30-minute call More case studies