This is one outlet's own report from Fortune — the article as it was filed.
AIPROPX ReportFortune · 3h ago
Inside Ikea’s big bet on humans in the age of AI
The countertop was a problem far too unruly for AI to solve.
When the customer called in to Ikea’s remote-sales center in Helsingborg to order a custom-cut slab for his kitchen counter, the dimensions he shared revealed it to be a lopsided hexagonal monster—no two sides the same length—with an electric stove in the middle. Could the Swedish home furnishings brand produce a countertop in such an irregular shape?
Melanie Lindell, the senior sales representative who had answered the call (in Swedish), ran the request up the chain, calling on Madeleine Barr, a senior sales specialist, who has been designing kitchens at Ikea for more than a decade.
“It’s a lot of angles,” Barr said, as she worked to puzzle it out. Soon Barr was back with good news: Ikea could make the dimensions work if the counter’s overhang was extended just a bit. Lindell smiled, evidently relishing the chance to guide the customer toward a solution—and a potential sale—that might otherwise have fallen through.
Lindell and her colleagues are at the center of a daring bet at Ikea : that in an era defined by automation, human interaction can be a revenue driver, not a cost to be stripped away. As the company has used AI to absorb routine customer service work, it has also retrained roughly 8,500 call center employees to handle more complex customer queries or work as sales-oriented design consultants.
Design advisors like Lindell use judgment, human chemistry, and friendly reassurance to turn calls that a bot can’t handle into conversations that have proved remarkably successful at driving sales.
Ikea’s 83-year-old business model is based on affordability—its founder Ingvar Kamprad was so frugal he reportedly reused tea bags—and it has maintained its low prices, in part, by stripping labor out of its system. Customers retrieve their own items from Ikea’s warehouses and typically assemble their Malm bed frames, Lack tables, and Hemnes dressers themselves.
But despite its already lean operations, Ikea didn’t join the AI-era rush to impose more austerity. When so many companies are measuring success in AI adoption by how many expenses they can cut and how many workers they can let go, Ikea is treating the technology as a way to free up human workers for higher-margin work that boosts its bottom line.
The Swedish furniture giant is quietly running a counterprogram to the standard AI story: Yes, it’s deploying automation to replace workers, but it’s also using the technology to make human skills more valuable.
And it’s providing one answer to the pressing question of what employers should do with workers whose roles are at risk from AI. Some 92 million jobs could be displaced because of AI, related technologies, and demographic shifts by 2030, the World Economic Forum estimates.
Five years ago, a worker in Lindell’s position—seated in a cubicle in an office park 65 kilometers north of Malmö, Sweden, a zip-up in Ikea’s trademark yellow draped over her chair—would likely have been fielding much more basic calls from shoppers. Among the most common: “What time does Ikea open?” and “Can I bring my dog into the store?”
That changed in 2021 when Ikea introduced its AI-powered customer service bot Billie, named after the retailer’s ubiquitous particleboard-and-laminate Billy bookcase, a staple of many a starter apartment. The bot could handle those repetitive questions with ease, and in its first two years it could assist 47% of customers who used the tool; now that figure is 74%.
Automating aspects of customer service is not unique to Ikea, but what happened next surely was: Instead of laying off the call center workers whose jobs the bot had partly taken over, Ikea retrained them to handle more complicated customer queries or to work as interior design sales advisors who help customers plan room redesigns and buy home furnishings. The two teams—resolutions and sales—operate from 24 remote-sales centers that cover all 31 countries where Ingka Group, the owner of most Ikea stores, has a presence.
The centers have been Ikea’s fastest-growing sales channel over the past three years, with year-on-year growth of between 15% and 20% annually. Last fiscal year, they accounted for 1.25 billion euros ($1.37 billion) in sales, up from 1.08 billion euros ($1.17 billion) the year prior. Meanwhile, Ikea says its in-house customer happiness score is now 89%, up from 60% prior to Billie’s rollout.
Ikea’s efforts to distinguish itself come at a crucial moment, as its flat-pack dominance is under pressure from Wayfair , Amazon , and design-forward challengers like Article. Ikea has responded to the pressure with an aggressive push on affordability . The chain’s biggest retail and franchising arms reported drops in revenue for fiscal 2025, even as customer visits and units sold edged higher. The bottom line: Every sale counts.
Brandon Mikula is one of the 8,500 workers Ikea retrained. Patrick Brown/Panos Pictures for Fortune
Those arms, Ingka Group (which owns and operates most Ikea stores worldwide) and Inter Ikea Group (which owns and franchises the Ikea brand globally), both laid off hundreds of corporate workers this spring to streamline operations and keep prices low. (None of the layoffs affected the remote-sales centers, nor were any attributed to AI.)
But the fact that these two companies that govern Ikea are private gives it the luxury of experimenting with price and people strategies that might spark shareholder blowback at a listed company. Perhaps that’s why it remains one of the few examples of a company that has successfully reskilled employees whose jobs have been automated—one of the biggest challenges of the AI age.
The advantage Ikea built with its AI-related reskilling initiative started from a deficit. Ikea was famously slow to embrace e-commerce . It had built its busine...
AIPROPX is an independent multi-source news index — we track, compare, and connect coverage from across the web into one place you won't find anywhere else.
