Transformation of JD Insurance Agent System: Architecture, Challenges, and Solutions

1. Introduction

JD is widely known for its e‑commerce platform, but behind it lies a dedicated team of JD insurance agents who provide protection services rather than sell goods. The Insurance Agent Department focuses on delivering comprehensive technical and business support to agents, aiming to optimize services, create opportunities, and promote industry growth.

Agents act as a trust bridge between customers and the company; establishing clear regulations (the "Basic Law") is essential for maintaining this bridge, defining management, compensation, and incentive structures.

To improve service efficiency for agents and internal staff, a smart, automated, and personalized system architecture is required. The existing architecture no longer meets current needs, prompting a complete system reconstruction.

2. Basic Law Overview

The Basic Law is a set of management rules for insurance marketers (agents and staff), covering daily operations, promotions, penalties, and benefit distribution. It guides agents' career development, performance assessment, and compensation, directly influencing business stability and market competitiveness.

2.1 Core Content of the Basic Law

Daily Management: Standards for signing, changes, termination, and training to ensure team stability and professionalism.

Compensation & Benefits: Defines calculation methods for direct commissions, indirect commissions, non‑cash rewards, and benefits such as pensions.

Performance Assessment: Sets clear performance requirements and incentives for each career stage, evaluating business achievement rate, continuation rate, and team metrics.

2.2 System Business Flow

3. Challenges Facing the Basic Law

3.1 System Complexity

Data Processing Complexity: Large volumes of customer, policy, and performance data with diverse formats create significant technical challenges.

Customization Requirements: Three existing versions involve 36 commission items and 17 grades, demanding deep insurance knowledge and strong system customization capabilities.

Policy & Regulatory Changes: Frequent regulatory updates require a flexible, extensible system to stay compliant.

3.2 Business Complexity

Product Diversity: Multiple insurance products (life, property, auto) each have unique terms and rate structures.

Sales Channel Diversity: Agents sell via offline and online channels, each with distinct scenarios and customer needs.

Risk Management: The system must balance business growth with credit, market, and operational risk controls.

4. System Evolution

4.1 Full‑Scenario Architecture

4.2 System Refactoring Comparison

Before Refactor

After Refactor

Readability

Highly coupled, repetitive, chaotic code, hard to understand

Low coupling, duplicate removal, clear logic, reduced complexity

Flexibility

Low modularity, difficult to maintain and extend

Modular design, easy maintenance and extension

Reliability

0%

99.99%

Security

Missing error‑handling mechanisms

Automatic review, link tracing

Performance

Heavy real‑time data dependency, long response, 6‑hour execution

Pre‑load metrics, Drools rule engine, 30‑minute execution

5. Key Breakthrough Points

5.1 Modular Decoupling and Clear Responsibility

Component‑Based Design: Decomposes the system into autonomous components (rules, rewards, level management, performance, collaboration) to improve maintainability, extensibility, and reuse.

Configurability: Provides a UI for adjusting component behavior without code changes, simplifying rule customization and reducing operational costs.

Free Combination: Users can assemble component versions like building blocks to quickly create personalized Basic Law solutions.

5.2 Introduction of a Rule Engine

Drools is used to express commission and assessment rules via DRL, separating business logic from code and enabling visual configuration, automatic DRL generation, and intuitive rule inspection.

5.3 Multi‑Dimensional Execution & Data Version Isolation

The system can execute calculations per version, team, commission, or assessment component, with isolated data versions allowing repeatable runs and cross‑version comparisons, supporting accurate commission forecasting and performance insights.

5.4 Enhanced Development Verification Mechanism

5.4.1 Data Verification

Before business review, the system automatically checks commission calculations against predefined fee ratios, ensuring all payouts stay within reasonable bounds and preventing human errors.

5.4.2 Multi‑Dimensional Business Data Analysis

Integration with a BI reporting system provides drag‑and‑drop visual analytics, enabling rapid anomaly detection, cross‑department data sharing, and data‑driven decision making.

6. System Stability Construction

6.1 Alarm Mechanism

The system integrates UMP and MDC monitoring, configuring alerts for JVM, CPU, disk, connectivity, memory, load, TCP connections, network throughput, and retransmissions to ensure continuous, stable, and efficient operation.

6.2 Basic Law Task Execution Trace

Link Monitoring (PFinder): Automatic performance instrumentation for SpringMVC, JSF, MySQL, JMQ, etc., providing topology and cross‑service tracing for rapid bottleneck analysis.

Task Chain Monitoring: Critical tasks are tracked, with retries and escalation alerts (email/phone) to guarantee timely completion.

7. Summary & Outlook

Since the refactor, system accuracy has risen to 99.99%, supporting over 40,000 agents and changing their perception of the platform. Future work will focus on continuous performance, stability, and security improvements, embracing new technologies to drive further automation, intelligence, and business growth.

Scan to join the technical community