Overview:
The Swedish fintech company Klarna uses an AI assistant, built with OpenAI technology, to streamline and automate both its customer-facing shopping experience and internal customer service operations. The solution aims to simplify and accelerate product search, comparison, and customer support — reducing friction for end-users and lowering operational burden for Klarna.
By deploying this AI assistant across 23 markets and supporting more than 35 languages, Klarna handles large volumes of customer interactions (refunds, returns, payments, general inquiries) through conversational AI instead of manual support agents. The deployment extends beyond external customer service: Klarna also provided access to enterprise-grade AI tools for nearly all internal employees, enabling broad productivity and automation across functions.
Key Features:
Delivers conversational shopping experience: allows customers to describe what they want (e.g., “sneakers under $150”) and returns relevant product recommendations with direct links to Klarna’s price-comparison tool.
Handles multilingual customer service at scale: supports over 35 languages and operates 24/7 across 23 markets, enabling global customer coverage.
Automates routine support tasks: processes refunds, returns, payment queries, account issues, and general customer inquiries via AI — reducing reliance on human agents.
Accelerates issue resolution: reduces average resolution time from ~11 minutes (human support) to under 2 minutes for AI-handled interactions.
Provides internal enterprise-grade AI tools: rolls out AI (via “ChatGPT Enterprise”) for employees across departments (marketing, legal, communications, etc.) boosting internal productivity and supporting diverse functions.
Results & Impact:
Within the first month, the AI assistant processed 2.3 million conversations, covering roughly two-thirds of all customer service chats—equivalent to the workload of around 700 full-time agents.
Customer issue resolution time shrank from 11 minutes to under 2 minutes, significantly speeding up service delivery.
The repeat-inquiry rate dropped by about 25%, indicating improved first-contact resolution and reduced friction for customers.
Customer satisfaction scores for AI interactions are reported to be on par with human agents, indicating that automation did not degrade service quality.
Klarna estimates that this AI deployment could lead to as much as US$ 40 million in profit improvement in 2024.
Internally, adoption of generative AI among employees is high (with up to ~90% daily usage across many departments), supporting broader organizational productivity beyond customer service.
AI Technology:
AI Model Types: Generative AI / LLM-based (text) models from OpenAI.
AI Purpose: Automate user interactions (shopping, support), summarize/comprehend user requests, generate conversational responses, assist internal teams (productivity).
Application Type: Customer Service & Consumer Experience (external), Internal Operations & Productivity (internal), E-commerce / Retail support.
Target Users:
Online shoppers using Klarna’s services (consumers)
Customer service teams (support staff and management) at Klarna
Internal employees across departments (marketing, communications, legal, product, etc.) using AI for productivity
E-commerce operations / payment systems teams managing returns/refunds/workflows