/ Case study · Employer
Self-checkout in a WeChat mini-program: store queues from 30 minutes to 2
Designed and built a self-checkout WeChat mini-program for retail stores, integrating product, payment and store systems so customers could skip the queue.
- 30 to 2 minStore checkout wait time
The problem
At peak times, customers in stores waited up to 30 minutes at checkout. Adding tills wasn’t scalable, and most customers already had the app they needed in their pocket: WeChat.
Constraints
- Had to run inside WeChat’s mini-program platform and its rules.
- Integrates with product catalogue, pricing, payment and store systems.
- Exit checks had to stay simple for store staff.
Architecture
- A WeChat mini-program handles scanning, cart and payment.
- Backend APIs call the product/pricing service, the payment gateway and store/order systems.
- At the exit, store staff verify each purchase by checking the receipt and handling the security-tag step.
Key decisions
- Meet customers where they already are (WeChat) instead of shipping another app.
- Keep the payment flow on the platform’s native rails to minimise friction and trust issues.
How it’s evaluated
- Checkout wait time at peak, before and after launch.
Outcome
Store checkout wait time fell from 30 minutes to about 2 (30 → 2 min).
What I’d do next
Pair it with demand forecasting for staffing, and use the same integration pattern for AI-assisted product search inside the mini-program.
Stack
- WeChat Mini Program (WXML)
- Java (Spring)
- Go
- REST APIs
- MySQL
- Redis