Manufacturers ramp AI spend as only 2% fully embedded: McKinsey

COO100 captured 101 responses from $1b+ revenue operators, surveyed in June–July 2025.

Manufacturers are planning a sharp increase in spending on digital and artificial intelligence, even as most remain far from scaling AI across their operations, according to McKinsey.

The study found that whilst about one-third of companies spent less than 1% of cost of goods sold (COGS) on digital and AI over the past five years, only 7% plan to keep investment at that level going forward.

By contrast, 93% expect to spend more, including almost one-third planning to allocate at least 5% of COGS.

Despite the planned increase in investment, McKinsey said around two-thirds of companies are still at the exploration or targeted implementation stage for AI, and only 2% reported that AI is fully embedded across all operations.

COOs said they are prioritising AI’s impact on production capacity, labour productivity, quality, and end-to-end visibility, with most planning to scale a focused portfolio of five to 12 use cases by 2030.

The report highlighted a risk that companies may be underinvesting in the enablers needed to scale AI effectively. Cultural change and workforce reskilling were cited as the most common barriers, alongside constraints in data quality and IT or operational technology systems.

McKinsey also pointed to a widespread lack of AI-specific performance indicators, noting that companies with clear KPIs were significantly more likely to meet or exceed their targets.

McKinsey said COOs need to treat AI as a performance agenda rather than a series of pilots, with clear ownership of value delivery, KPI-linked targets, and dedicated funding.

Case examples cited in the report showed improvements in areas such as equipment effectiveness, labour productivity and production cycle times where AI was deployed at scale, often using a hybrid build–buy–partner operating model.

The findings are based on McKinsey’s COO100 Survey, which gathered 101 responses from operating executives at manufacturers with at least US$1b in revenue.

The survey was conducted between June and July 2025, with respondents primarily from North America, Western Europe, and the Asia–Pacific.