Klarna has filed paperwork to list on the New York Stock Exchange (NYSX) following an extensive cost-cutting exercise that has transformed its approach to customer experience (CX), employee experience (EX) and customer service.
Klarna confirmed on March 14 that it has filed a registration statement with the US Securities and Exchange Commission. The number of shares to be offered and the price range for the proposed offering have not yet been determined. Regardless, the listing is set to be the largest by a Swedish company in the US since Spotify listed in 2018.
As a buy now pay later (BNPL) service, Klarna is available to the customers of thousands of brands, from Airbnb and Shein, to luxury retailers like Neiman Marcus and Bergdorf Goodman. It announces new payment partnerships almost weekly and, as of last month, its BNPL services are also available to businesses that rely on J.P. Morgan Payments for payment processing.
As of December 31, 2024, Klarna’s network comprises more than 90 million active consumers and more than 675,000 merchants in 26 countries, and it facilitated US$105 billion of gross merchandise volume in 2024. Driven by global growth in the BNPL market – which is expected to reach $160bn by 2032 – the value of Klarna increased 24 percent in 2024. Between mid-2020 and 2021, Klarna conducted three funding rounds which saw its valuation rise from $5.5bn to $46.5bn.
Despite this strength, it has posted losses in the past; $1bn in 2022 and $241mn in 2023.
In this context, it makes sense the company is looking to trim its losses ahead of listing on the NYSX. The role of artificial intelligence (AI) and automation cannot be understated in this work. Although Klarna made its name in BNPL, the business now describes itself as an “AI-powered payments and commerce network and shopping assistant” proving that, like many organizations today, its strength is driven by its use of technology just as much as the reach of its brand.
Using AI to improve a company's profitability
As CX Network covered last year, Klarna stopped hiring human workers as it looked to boost profitability ahead of its planned IPO.
The table below illustrates how AI has been used to replace human workers over the last three years. CEO Sebastian Siemiatkowski told the Financial Times last year the company could run on 2,000 employees as its AI use ramps up further, predominantly driven by in-house builds.
Klarna works closely with OpenAI, using its large language models (LLMs) to drive better CX and EX and is also using AI to build its own automation technology.
Internally, more than 87 percent of Klarna employees were using generative AI in their work as off May 2024, across communications (93 percent adoption), marketing (88 percent adoption) and even legal (86 percent) where company lawyers use ChatGPT Enterprise to create the first draft of common types of contracts, reducing the time it takes to draw them.
Klarna says the high adoption rate – particularly among non-technical teams – is driven by the use of its “bespoke internal AI assistant”, Kiki, which handles around 2,000 inquiries a day.
As illustrated in the table, this approach has drastically reduced Klarna’s human workforce. As announced in August 2024, the plan for the “coming years” is to reduce headcount even further.
Together, the changes allow Klarna to reduce financial losses and operating costs and boost its valuation for an IPO.
Compared with 2023 figures, Klarna said in mid-2024 that its average revenue per employee had increased 73 percent year on year. This was partly driven by an AI chatbot, launched in January 2024, that now performs the work of 700 human employees across Klarna’s customer service suite.
AI is not only replacing employees. Siemiatkowski said last year that as a result of its AI strategy, Klarna is dropping Salesforce, Workday and many of the other software-as-a-service (SaaS) providers it was using for customer relationship management (CRM), human resources and payroll services.
Klarna’s evolution: From BNPL provider to AI pioneer
Klarna’s chief product and design officer, David Fock, said last year that Klarna is “enlisting AI to challenge the incumbents”, and the pace at which this is happening shows no sign of letting up.
In September 2024 Klarna unveiled more features for its OpenAI-powered AI assistant.
The bot now helps customers to search for specific products or brands and access key product insights within a single chat interface. It can compare products or different categories of products, to inform customer shopping decisions. The bot can recommend products based on customer preferences, enhancing the relevance and satisfaction of a shopping experience, and even reverse search the websites that offer Klarna as a payment method to find the best price for millions of products, across thousands of merchants, along with stock, delivery and cashback info.
In February this year, Klarna said the assistant had handled 2.3 million customer conversations – two-thirds of Klarna’s customer service chats – in its first month of being live in 23 global markets.
The assistant isn’t just handling volume. Klarna’s first contact resolution (FCR) rate is also strong, with the assistant delivering a 25 percent reduction in repeat inquiries and allowing customers to resolve issues in less than two mins compared to 11 mins previously.
Although full year 2024 figures are yet to be released, it’s estimated the assistant will have delivered a $40mn profit improvement to the company for the full year.
Quick links
- Can Honor’s deepfake detection restore customer trust in AI?
- Shein IPO hits another hurdle as company value slides
- Trump versus Consumer Protection
[inlinead-1]