When I hear the word “factory,” I picture the traditional scene: long production lines, noisy machines, and workers moving among them. But the more I read about modern industry, the clearer it becomes that this picture is no longer enough. Today, factories are run by data as much as by machines—and sometimes make more money after the sale than from the sale itself.

I found a clear account of this shift in Deloitte’s “2026 Manufacturing Industry Outlook,” published in November 2025. Although the report focuses on the U.S. market, the trends it identifies extend beyond its borders.

Some necessary context

The report paints a challenging picture of U.S. manufacturing in 2025. The manufacturing purchasing managers’ index remained below 50 for most of the year, signaling contraction, while costs rose and employment and spending on factory construction fell. The share of manufacturers who cited trade uncertainty as their biggest concern also rose from 56% to 78%, according to surveys by the National Association of Manufacturers in the United States.

Against this backdrop, the report identifies five trends it sees as key to staying competitive.

Trend one: smart manufacturing

Smart manufacturing connects machines and production lines to sensors, data, automation, and cloud computing, allowing a factory to see what is happening inside it in real time and make decisions accordingly.

According to a Deloitte survey of 600 manufacturing executives, 80% plan to allocate 20% or more of their development budgets to smart manufacturing. Some 92% see it as the main driver of competitiveness, up from 86% in 2019. Companies that have implemented it have reported average gains of 10% to 20% in production, 7% to 20% in employee productivity, and 10% to 15% in previously idle production capacity.

The report also notes the rise of what is known as “physical AI”—AI that controls real-world machines and robots. Some 22% of companies plan to use it within two years, compared with 9% today. It also expects a growing role for “agentic AI,” which carries out multistep tasks such as suggesting alternative suppliers, preparing shift handover reports, and speeding up equipment repairs.

Trend two: supply chains under pressure

According to the report, U.S. manufacturers expect their input costs to rise by an average of 5.4% over the next year. In response, many companies have stockpiled materials, restructured their sourcing, modeled tariff scenarios, and renegotiated with suppliers. The report recommends using AI to monitor direct suppliers and their suppliers, assess the impact of any disruption, and suggest alternatives—while leaving the final decision to people.

Trend three: a new wave of investment

Tax incentives and policies to reshore manufacturing are driving a wave of investment in U.S. factories. In the semiconductor sector alone, private investment commitments had exceeded $500 billion by July 2025. This could triple U.S. production capacity by 2032 and create more than 500,000 jobs.

One detail that caught my attention was that major manufacturers of electrical equipment—such as transformers, switchgear, and power-generation systems—are signing multiyear agreements, and some have already sold out their production capacity for several years ahead. New factories and data centers need electricity before anything else.

Trend four: profits after the sale

In my view, this is the report’s smartest insight. After-sales services, such as spare parts, maintenance, and service contracts, generate profit margins for industrial manufacturers that are more than twice those of equipment sales themselves.

The report sees a role for AI here, too: detecting component wear before a breakdown, ordering spare parts, scheduling maintenance, and assessing warranty claims—with the final decision still in human hands. In other words, selling a machine is no longer the end of the relationship with the customer; it is the beginning.

Trend five: skills before machines

More than a third of the executives surveyed said that upskilling workers in smart manufacturing is their biggest concern. The report suggests three ways to close the skills gap: developing talent internally, recruiting specialists from outside, and using temporary or outsourced workers for non-core roles. AI can also preserve the tacit knowledge of experienced workers and turn it into operating procedures that speed up training for new hires.

The figure that stood out to me most: people are expected to continue performing more than 81% of the hours spent on tasks in the manufacturing sector.

When Germany coined the term “Industry 4.0” in 2011, it was describing factories where machines connect and exchange data. Years later, it seems the smart factory has not removed people from the equation; it has changed their place in it.

That may be why the best description of a smart factory is not a factory without people.

A more realistic description, in my view, is a factory that knows more: it knows when a machine is about to fail, where risks in the supply chain are coming from, and where it can make money after a sale. People remain at its heart, but with different skills.

That is why I think Deloitte’s report is worth reading—not simply as a matter of U.S. interest, nor as a ready-made formula for every factory, but as a window onto the broader transformation of industry. It is a transformation we are also experiencing in the Kingdom through the “Future Factories” program, which aims to turn 4,000 factories into smart facilities:

From a factory run on experience alone to one run on data and experience together.

Disclaimer: This article is intended to explain and simplify the subject for readers. The information it contains is drawn from published sources and has been adapted and presented in accessible language. It does not represent an official position or statement by any organization. The information may be updated after publication, so readers are advised to consult official sources for full details.