Health Plans: Your BI Tells You MLR Moved. Can Your AI Tell You Why?

A health plan CFO notices a higher-than-expected medical loss ratio (MLR) after reviewing monthly financials, but traditional business intelligence only confirms the variance without explaining why. Spreadsheets and manual reconciliation delay insights, leaving leaders without actionable answers. AI aims to transform this process by not just identifying the variance but diagnosing its drivers, such as utilization, unit costs, or service mix. The goal is to shift from reactive reporting to proactive investigation, where executives can ask follow-up questions in plain language and receive immediate, context-rich responses.

For decades, health plan finance teams relied on rigid dashboards and analyst-driven reports to track metrics like MLR, a metric defined as the share of premium revenue spent on medical care. Calculating MLR requires combining claims, pharmacy data, rebates, and other components, then segmenting by line of business or market. This complexity often forces teams to reconcile conflicting definitions across systems, delaying accurate insights. AI promises to streamline this by interpreting data in real time, but only if the underlying information is consistently structured and understood within the plan’s operational context.

Databricks and Abacus are collaborating to address this challenge by merging Databricks’ data and AI capabilities with Abacus’ payer-specific data foundation. Abacus organizes health plan data—claims, eligibility, providers, contracts, clinical records, and financials—into a unified structure that reflects how the business operates. This foundation ensures consistent definitions and relationships, which are critical for AI to generate reliable insights. Without this context, even advanced AI systems may produce misleading or incomplete answers, undermining trust in their recommendations.

The partnership shifts the focus from data access to data understanding, enabling finance leaders to ask nuanced questions about MLR or other metrics and receive answers grounded in the plan’s unique business logic. For example, a CFO could inquire why MLR is above budget, then drill into specific populations, services, or providers driving the variance. This approach breaks down silos between executives, who understand the business, and analysts, who understand the data, allowing both to work from a single source of truth. The result is faster, more confident decision-making across financial and operational domains.