Design and Implementation of JD.com Data Construction Platform for Testing Efficiency

The platform was created to address the growing testing challenges caused by rapid development of JD.com’s home‑delivery and instant‑purchase services, where manual test data preparation consumed over 30% of the testing cycle and was labor‑intensive.

Its goals include reducing the time required to generate diverse test scenarios, improving the technical skills of business testers, and consolidating data construction capabilities into a single system accessible to various product, research, and testing teams.

Built on a mature Springboot + Mybatis + Vue stack, the system integrates account and configuration services, supports environment switching, permission control, and data‑tracking. It provides a unified interface for creating both static (e.g., merchant, store, product) and dynamic (e.g., delivery time, freight) test data.

Key functional modules include:

One‑click order creation: Simulates real user order submission by capturing API calls and parameterizing only necessary inputs, ensuring generated data matches actual orders.

Order status flow: Enables rapid transition of orders through more than 95% of status nodes with a single click, dramatically cutting the 20‑minute manual process to seconds.

Coupon creation: Supports one‑click generation of multi‑type, multi‑channel coupons, reducing the lengthy manual entry required for marketing tests.

Permission restrictions: Differentiates between basic and sensitive functions, applying whitelist checks and user‑pin validation to prevent accidental manipulation of production data.

Since deployment, the platform has delivered measurable benefits: one‑click order creation reduced processing time from 20 minutes to 20 seconds, saving over 1,600 hours; status‑change tools saved more than 127 hours; and overall test cycle time decreased by roughly 30%.

Future plans aim to expose data‑construction APIs to external services and integrate with automation pipelines, further enhancing testing efficiency and expanding the platform’s capabilities beyond post‑order scenarios.