Customer Service Automation
Yum China had reached the limits of small-model online customer service and needed more flexible dialogue handling for delivery-status, ordering and other high-volume restaurant-service scenarios.
Reported impact
Yum China raised answer quality and self-service resolution rates while lowering the cost of handling large-scale digital customer-service traffic across restaurant brands.
Yum China upgraded its digital customer-service stack with Baidu AI Cloud so restaurant support could move beyond rigid FAQ flows. The new system is built around Baidu AI Cloud Keyue and adds large-model intent recognition and multi-turn dialogue to existing knowledge-driven support. In operation, requests from the app, mini-programs and food-delivery platforms are routed into the AI service layer, which identifies intent, keeps the context of the conversation and responds more flexibly than the earlier small-model system. Complex scenarios such as催单 are where the change matters most: the assistant now uses context and user state to produce a specific resolution rather than a generic scripted answer. This makes the platform a conversation engine for very high-volume restaurant-service interactions, one that both improves customer self-service and lowers handoff to human agents across multiple digital channels.
Consumer requests from apps, mini-programs and delivery platforms enter the Keyue-based service layer, where intent recognition and multi-turn dialogue generate responses and route unresolved issues away from manual service channels only when needed.
Keyue uses large-model intent recognition and multi-turn dialogue to upgrade the customer-service conversation experience (original: 客悦通过融合大模型的意图识别、多轮对话等能力……升级客服对话体验).
Robot-answer accuracy reached 95% and self-service resolution reached 90% (original: 机器人答复准确率达到95%……消费者自助解决率提升至90%).
The AI customer-service system is already online across the app, mini-programs and delivery platforms and handles more than 150,000 conversations per day (original: 百胜中国AI智能客服系统已经在APP、小程序、外卖平台等多个渠道上线，每天为消费者提供超过15万次的沟通服务).
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
OpenAI Sample Modified · Oct 16, 2026 Grupo Boticário Grupo Boticário needed to support a complex multi-brand reseller and franchise ecosystem across many channels while reducing repetitive operational work and preserving the human relationships that drive sales and… 100% of conversations generated in appchat and WhatsApp Enterprise covered since November 2025 OpenAI foundation models · WhatsApp Enterprise · appchat 
Google Cloud Sample Modified · Oct 9, 2026 Central Retail Food Replaced ~2,000 monthly manual Excel reports with a single source of truth on BigQuery through Looker dashboards Central Retail Food unifies its ecosystem on Google Cloud, moving from fragmented data to conversational… Addresses 50% to 60% of customer service queries with automated chatbots built on Conversational Agents BigQuery · Gemini · Gemini platform Google Cloud Sample Archived · Oct 8, 2026 Smyths Toys Smyths Toys needed to handle growing customer-service volume across email, web, and social channels while giving AI agents access to live order, shipping, refund, and product data without slowing or destabilizing core… Resolved more than 300,000 customer emails BigQuery · Gemini Enterprise · Gemini Enterprise Agent Platform 
