Design and Implementation of JD's 01 Payment Platform: Unified Payment Service Layer and One‑Stop Integration

JD's retail payment flow includes search, product detail, cart, checkout, order, and payment, with payment being the critical final step. The legacy checkout system suffered from severe code decay, poor scalability, and slow first‑contentful‑paint due to heavy resource loading.

To address these issues, the retail payment team created the high‑availability, high‑scalability 01 Payment Platform, which combines a unified payment service layer and a one‑stop integration platform, delivering a single‑checkout solution across native, H5, and mini‑program channels.

The unified payment service layer aggregates core payment capabilities, abstracts each payment method into independent components (e.g., JD Pay, WeChat Pay, digital RMB), and exposes SaaS and PaaS layers to support Android/iOS SDKs, PC/M‑site payments, and mini‑program plugins.

The SaaS layer enables multiple checkout forms (full‑screen, popup, M‑site, regulatory) and provides cross‑channel payment capabilities via native SDKs and web plugins.

The PaaS layer refactors the old SOA system, offering a unified service, one‑stop integration, and intelligent monitoring, reducing integration complexity and supporting low‑cost, extensible payment, risk control, and personalization extensions.

Security is enhanced by isolating core payment logic from external business logic, achieving tenant isolation and safer innovation.

The platform also includes a self‑developed intelligent monitoring system that unifies data processing, supports UDP/Kafka reporting, and offers configurable, SQL‑based analytics (funnel, attribution, path, hotspot, event, correlation) with visual dashboards and export capabilities.

The one‑stop integration platform streamlines onboarding for business units, providing secure, stable, and customizable access along with comprehensive monitoring and alerting.

During the 618 promotion, the 01 Payment Platform replaced the legacy SOA system, handling increased traffic without functional loss, delivering higher performance, better user experience, and stable architecture.

It supported over 265 online business scenarios across iOS, Android, H5, and mini‑programs, enabling rapid, self‑service integration and reducing development effort.

Future plans include expanding intelligent monitoring across the full payment chain, providing multi‑dimensional data analysis, and decentralizing core services to allow business units to configure and run A/B experiments autonomously.