Customer Service Automation
As order volume and service complexity grew, Deppon Express needed to lower communication cost, improve response speed, and deliver consistent service across many digital channels without overloading human agents.
Reported impact
Deppon Express created a consistent omnichannel service model for logistics support, reduced manual handoffs, and improved customer-response efficiency while continuing to route complex cases to live agents.
Deppon Express built an online customer-service platform with Baidu AI Cloud to create a dual-track service model that combines automation with human escalation. The implementation paired Baidu Intelligent Customer Service, Baidu Keyue, and Deppon's online customer-service IM system so the company could standardize logistics-domain scripts, accumulate reusable Q&A knowledge, and expose the same service logic across more than ten customer touchpoints. Incoming questions from mini programs, the website, the mobile app, Alipay, Douyin, and other channels flow into the intelligent layer first, where routine intents are answered automatically and unresolved issues are handed to live agents with a shared knowledge context. The setup gives Deppon one service backbone for consistent logistics answers, faster response times, and lower transfer-to-human rates while preserving human support for complex requests.
Questions from WeChat mini programs, the website, the app, Alipay, Douyin, and other service entries are routed into the intelligent customer-service layer. Routine intents are answered from Deppon's logistics knowledge base, while harder cases are transferred into the IM system for human follow-up with the same script and context foundation.
Deppon deployed Baidu's intelligent customer-service system to build an 'AI + human' dual-track service model (original: 通过部署百度智能客服系统，构建"AI+人工"双轨服务体系).
The intelligent platform combines Baidu Keyue with Deppon's online customer-service IM system (original: 德邦快递通过建设智能客服平台，搭配百度客悦与在线客服IM系统).
The project built a logistics-industry-specific script system and connected more than ten service entrances including WeChat, Alipay, and Douyin (original: 搭建物流行业专属话术体系，统一对接微信/支付宝/抖音等10余个服务入口).
The transfer-to-human rate was reduced to 10% (original: 转人工率降低至10%).
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
Baidu AI Cloud Sample Oct 13, 2025 Shandong Port Group (山东省港口集团有限公司) Shandong Port Group needed a better way to serve many kinds of logistics customers because information was fragmented, multimodal transport coordination was weak, and inconsistent service processes limited logistics… Intelligent Customs Clearance · Intelligent Tracking · Port Encyclopedia 
In-house Sample Est. · Jul 16, 2026 Poste Italiane Poste Italiane needed to use its large, distributed data estate and national omnichannel service network more effectively so employees could respond faster during customer interactions, customers could access services… 25 million daily customer interactions AI-supported telephone booking service · Cloud and Edge Computing · Poste Italiane single app 
OpenAI Sample May 6, 2026 Uber Every day, millions of people rely on Uber to book rides, order meals, send packages, and earn flexibly. Behind every tap is a complex real-time marketplace shaped by traffic, weather, airport arrivals, local events,… Every day, millions of people rely on Uber to book rides, order meals, send packages, and earn flexibly. Behind… Realtime API · Uber Assistant 