Back Issues/Search Home → Calendar → Archive → RSS → Subscribe → Services → Vendors → Blog → Current Issue → Popular → Newsletters (LinkedIn, Substack)

All issues › Volume 344, Issue 2 › IT Vendor News › MongoDB

How to Model an Insurance Policy without Joining Dozens of Tables

MongoDB, Friday, November 13th, 2026

MongoDB explains modeling insurance policies as documents around the policy lifecycle, using an embed-or-reference rule.

MongoDB argues that insurance core systems slice a single policy across dozens of relational tables that must be joined on every read and altered on every product change, and cites Manulife replacing more than 500 core tables with one document per policy.

The recommended approach models data around the policy lifecycle and what is read and written together, reducing design to one decision: embed or reference.

Coverages, insured vehicles and event history are embedded in the policy, and reserves and payments in the claim, so updates are atomic.

Customers, policies and claims are kept in separate collections linked by ID because they have their own lifecycles.

more →  ·  More from MongoDB →