For the past two years, many AI stories have followed the same script: Automate work, reduce headcount, and cut costs. IKEA chose a different path, and its results offer an important lesson for leaders navigating the AI era. Even though this story is years old, it holds key insight into grappling with the workforce impacts of AI.
What Happened At IKEA?
Back in 2021, Ingka Group, IKEA’s largest retailer, deployed an AI-powered customer service chatbot (“Billie”) to handle routine customer inquiries. The bot handled 47% of customer calls, saving Ingka €13/$17.1 million in savings. Rather than eliminating approximately 8,500 employees whose work had been affected by automation, the company analyzed customer demand and found an unmet need: Customers wanted more help designing their homes. IKEA reskilled and redeployed those employees into a new interior design and home-furnishing advisory service. That new business line generated approximately €1.3/$1.7 billion in revenue in 2022, representing 3.3% of total revenue and expected to continue growing.
When AI Replacement Goes Wrong
IKEA’s approach stands in stark contrast to a growing number of organizations that discovered replacing people with AI isn’t as simple as it appears.
Klarna, once celebrated for using AI to perform the work of hundreds of customer service agents, later acknowledged that service quality suffered. The company began reinvesting in human-supported customer service after realizing that AI struggled with complex, sensitive customer interactions.
Ford recently rehired more than 300 experienced engineers after automated quality systems and AI tools failed to deliver the expected results. The returning experts helped identify problems AI missed, train younger employees, and improve the AI systems themselves.
Commonwealth Bank of Australia admitted it “did not adequately consider all relevant business considerations” before replacing 45 customer service agents with AI. Once the chatbot was implemented, remaining employees said that their level of work spiked.
IBM laid off hundreds of HR professionals, replacing them with AI agents. While the agents successfully handled about 94% of routine HR requests, they struggled with complex employee situations requiring judgment, empathy, and nuanced decision-making. The company subsequently increased hiring in areas where human capabilities remain essential, and plans to triple its US entry-level hiring across all business units this year.
The Takeaway
The organizations creating the greatest value from AI are not asking, “Which jobs can we eliminate?” They’re asking, “What higher-value work can people do once AI removes the repetitive tasks?” Think about it: As all organizational leader move toward using AI across their business, their people — with all our creativity, nuance, judgment, musings, and connections to each other — will bring differentiation in the AI-sameness we’re all heading towards.
Forrester clients can talk to us about how to identify high-value use cases for AI, how to create an AI-ready culture and an AI-ready workforce, and the workforce data you’ll need to identify existing talent and redeploy them.
