Order Status and Shipment Visibility Agent

Cellbes, a Scandinavian fashion ecommerce retailer, deployed an AI chatbot integrated with order tracking systems to address an overwhelmed support team inundated with repetitive delivery questions. According to a case study published by Kindly, the chatbot provider, the system now handles 77% of incoming chats in the Swedish market with a 95.6% understanding rate, meaning the bot correctly interprets and responds to nearly all interactions. The retailer reports that the number of questions handled by human agents has decreased, and because Cellbes outsources customer support, the reduction in contacts translates directly to lower costs. The implementation took weeks rather than months, and Cellbes plans to expand the chatbot to additional European markets.

At a larger scale, Klarna, the Swedish buy-now-pay-later provider, launched an OpenAI-powered AI assistant in early 2024 that handled 2.3 million conversations in its first month, representing two-thirds of all customer service chats. According to Klarna's published results, the assistant reduced average resolution time from 11 minutes to under two minutes, decreased repeat inquiries by 25%, and projected a $40 million profit improvement for 2024. However, Klarna's experience also illustrates the limitations of an AI-only approach: by mid-2025, CEO Sebastian Siemiatkowski acknowledged that cost had been "a too predominant evaluation factor," resulting in lower quality for complex queries, and the company began rehiring human agents to handle nuanced interactions alongside the AI system. This hybrid recalibration underscores that AI order status agents perform best when deployed for high-volume, routine inquiries while preserving clear escalation paths to human support for exceptions and emotionally sensitive cases.