Overview
A reinsurance firm processed 2,400+ incoming statements monthly from dozens of insurance carriers and financial institutions. The handling was entirely manual — collect, identify, classify, file. The team consistently fell behind before major tax deadlines and regularly hired seasonal temporary employees to absorb the load.
Building the automation was the easy part. The constraint shifted: making the organization actually trust it, use it, and depend on it. That took longer and required different work.
The Problem
The constraint wasn't staffing. The constraint was that manual document handling at that volume was not sustainable. And building automation was only the first problem.
The initial system was technically functional but optimized for developers, not operational users. Usability gaps, reliability concerns, and unfamiliarity with the learning mapper created friction. The organization resisted depending on something it didn't fully trust.
The real work was making the system reliable enough, understandable enough, and embedded enough that daily use became the default.
The Approach
I approached this as an adoption problem, not just a technical problem:
- Technical foundation: Three web crawlers collecting 2,400+ statements monthly from highest-volume providers; drag-and-drop processing tool with learning mapper achieving 68% automated filing rate
- Usability focus: Interface improvements targeting operational (non-technical) users
- Reliability investment: System stability to earn trust
- Training and change management: User education and organizational embedding
- Sustained engagement: 3+ years of continuous improvement and support
What Was Built
- 3 web crawlers — For automated document collection
- Drag-and-drop processing interface — For operational users
- Learning mapper — Achieving 68% automated filing rate
- Usability and reliability improvements — For operational users
- Training program — And change management strategy
- Sustained adoption support — Over 3+ years
Outcomes
The organization adopted it, trusted it, and built dependency on it — and that's what produced the outcomes. Team stopped hiring seasonal temporary staff before tax deadlines. First tax season completed ahead of deadline. Additional teams requested the capability. Pattern extended into accounting workflows and automated government form filing.
Why This Matters
This project demonstrates:
- Workflow automation — Technical system design and implementation
- Operational design — Making systems work within organizational constraints
- Adoption strategy — The difference between building and embedding
- Sustained engagement — Long-term operational change management
The key insight: Automation value only materializes when the organization genuinely depends on the system.