Overview

Cloverleaf Analytics provides an analytics platform for property & casualty insurers. They run dedicated Snowflake environments for customers and offer OEM analytics tools inside their product. Cloverleaf wanted to add natural-language question answering as a complement to their existing analytics experience, giving customers a faster, more intuitive way to get answers than static dashboards, without rebuilding their stack.

The Challenge

Natural language approach wasn’t production-ready: The existing set-up relied on prompts sent to an external LLM which was slow, low-context, and sometimes nonsensical, which made it unsuitable for executive or customer-facing use.

Insurance complexity + strict security requirements: Cloverleaf needed support for complex underwriting logic, role-based access controls, and reviewed queries so answers were correct and governed per customer.

Embedded product economics mattered: The solution needed to map to Cloverleaf’s product strategy: monetization, time-to-value for customers, and differentiation to protect market share and reduce churn.

The Solution

WisdomAI Embedded integrated directly with Cloverleaf’s Snowflake-first architecture to deliver governed natural-language question answering inside Cloverleaf’s platform.

‍

What this enabled:

• Embedded natural-language Q&A inside Cloverleaf
• Accurate, context-aware answers vs. external LLM prompts
• Governed results across customer-isolated Snowflake stacks
• Insurance-specific logic + role-based access (RBAC) with reviewed queries
• New upsell revenue a differentiated customer experience
• Faster feedback loops and shorter customer engagements