AI companies making the most headway tend to organise their teams around products, customer journeys, and core business processes, according to a McKinsey report released in recent months. That structure, the report noted, reduces decision-making layers and spreads accountability more broadly across the business.
The document drew no straight line toward organisational achievement, noting instead that multiple routes remain open. Proponents cited in the report argue that organisations would benefit from leaner processes, fewer handover points, and the flexibility to shift resources as priorities evolve.
Four in five participants surveyed for McKinsey's 2026 State of AI report said they had noticed a lift in their own day-to-day output when using AI tools. Earnings before interest and taxes told a more restrained story: 37 per cent of respondents noted a lift in EBIT, and the proportion meeting the threshold for high performers stood at just 6 per cent.
The report found that just one in nine organisations had advanced to the point of fundamentally reimagining their workflows and the ways AI generates value within them.
According to the document, a well-functioning operating model allows organisations to reach decisions, exchange information, and align their work through structures carrying fewer layers of management, a reduced number of hand-offs between teams, and fewer processes dependent on manual effort. Coordinator positions had been expanding at a pace roughly one-and-a-half to twice that of frontline roles across the preceding five years, the report noted — a pattern pointing to deepening structural complexity within organisations. The document indicated that artificial intelligence holds the potential to streamline that complexity rather than add further layers.
Erik Brynjolfsson, an economist at Stanford, told McKinsey that narrowing the AI conversation to headcount cuts leaves a broader opening untapped: the chance to restructure how work itself is organised around what the technology makes possible.
The document identified a potential drawback: when automated systems absorb the routine tasks through which workers have traditionally built their capabilities, near-term output gains may leave future professionals without the hands-on grounding needed to exercise sound judgment.
A separate poll highlighted a gap between how prepared employees believed themselves to be and how thoroughly their organisations had invested in AI readiness. Survey data showed that roughly seven in ten respondents felt personally equipped for AI adoption, yet fewer than three in ten leaders reported their organisations had the foundations in place to navigate an agentic AI transition.
The document called for structured guidance on AI deployment and clear delineation of staff responsibilities. One life sciences firm operating worldwide was highlighted as an example — its internal guidelines specify that artificial intelligence serves as a tool to augment work rather than displace the judgement, creative input, or independent oversight that employees bring to their roles. Under those guidelines, when AI tools contribute to a piece of work, the person who produced it retains full responsibility for the outcome.
Where AI stands to deliver the greatest returns, the findings suggested, hinges less on how many automated agents a business deploys or how broadly it eliminates manual tasks than on other factors. The report described embedding AI across the full span of how work is structured, how talent grows, and how leadership functions as part of a broader rethink of how organisations generate value.
