Industrial companies should protect R&D during downturns. Cutting engineering to save short-term profit is a deceptive reflex and a deferred technological bankruptcy, argues Bruno Bouygues, Chairman and CEO of GYS, a French manufacturer of machine tools.
Should companies cut engineering in a downturn?
No. In a context of shrinking markets, the pressure to protect margins is intense, and engineering budgets are an obvious target. Cutting engineering to save immediate operating profit is a deceptive reflex. It is not a saving. It is a deferred technological bankruptcy in the face of rivals who, for their part, are accelerating.
The logic is simple. Savings in engineering show up in this year’s results. The loss shows up later, when a competitor launches a better machine and the company has no product ready to answer. By then, rebuilding the capability costs far more than the budget that was cut.
That is why the leadership team must protect R&D efforts from short-term economic turbulence. It is one of the few decisions that only the top of the company can make, because it means accepting a weaker quarter to secure the next five years.
Why does speed to market matter so much?
Windows of commercial opportunity are shrinking. They close at a pace dictated by global competitors, notably Asian ones, able to model their iteration cycles on those of consumer tech. A machine that arrives late may arrive obsolete.
Speed to market has therefore become a core measure of leadership. The mission of the head of a machine manufacturer is no longer just to allocate capital, arbitrate budgets and control margins. It is to orchestrate the speed to market of highly complex solutions. A company with excellent engineers but slow decisions still loses.
How is industrial production changing?
For decades, excellence in INDUSTRIAL PRODUCTION meant two things: external growth through acquisitions, and operational excellence through lean manufacturing, line automation and permanent cost compression. These tools remain necessary. They are no longer a sufficient shield.
The most automated factory in the world is pointless if it produces machines the market already considers obsolete. The frontier of value has moved. It is no longer found only on the production lines. It has moved to the design offices, the laboratories and the teams that conceive tomorrow’s offering.
The machine itself is becoming software. It evolves after installation, extends its capabilities through updates, generates data and plugs into broader customer ecosystems. As a result, obsolescence is accelerating, and the engineering teams that create new generations of machines are the ones that decide whether the factory has anything valuable to build.
Why is complexity the real bottleneck?
Industrial machines are no longer simple electromechanical assemblies. They have become cyber-physical systems combining power electronics, real-time firmware, connected sensors, cybersecurity and artificial intelligence. These disciplines, long kept in separate silos, must now integrate seamlessly.
The slightest misstep in the software architecture, or in the choice of a critical component, can paralyze months of engineering. Protecting engineering capacity is therefore not a luxury. It is protecting the company’s ability to launch products at all.
What the tech industry now calls “physical AI”, artificial intelligence embodied in physical systems and able to perceive and act on the real world, adds to this challenge. It is already the ground on which differentiation is being built. Skills in this area take years to develop and cannot be rebuilt quickly after a round of cuts.
What is the French State signaling?
On July 10, 2026, the French State launched the “Défi Flagships” call for projects under France 2030. It aims to support innovative subsystems for robotics, drones and smart equipment.
This public choice says something essential: French industrial competitiveness will no longer be determined by production volumes, but by companies’ ability to integrate electronics, software and embedded intelligence into their machines faster than their competitors. A company that cuts the engineering teams who do this integration is moving in the opposite direction to the market, and to public policy.
Why are acquisitions not a substitute for R&D?
Some leaders respond to slower organic growth by buying growth. The approach has limits. Buying a competitor usually brings business volume, a customer portfolio and assembly capacity. All too rarely does it buy a technological lead.
Without an architectural vision shared by the entire leadership team, multiplying acquisitions often leads to piling up invisible technical debt: heterogeneous platforms and disparate generations of machines, costly to maintain. Engineering capacity is then needed more than ever, to simplify and unify what has been bought. Cutting it at that moment compounds the problem.
What must the leadership team do?
Protecting R&D is one part of a broader change in how executive teams operate. Five actions stand out:
- Understand technology well enough to challenge engineering teams, whether they are mechanical engineers, electronics engineers, automation specialists or software developers.
- Break down silos between mechanical engineers, electronics engineers and software developers.
- Simplify product architectures so that products launch faster and cost less to maintain.
- Treat technology as a core function, not a support function.
- Decide, as an executive team, where future margin will come from: hardware, software or connected services.
This shift of attention is still marginal on the agenda of many executive teams. It is becoming a condition for survival and competitiveness.
What does this mean for the CEO?
The CEO does not replace the chief technology officer or meddle in every line of code or wiring diagram. The CEO becomes a chief engineer in spirit: a leader whose technological understanding is intimate enough to ask the uncomfortable questions, decide the major trade-offs and set investment priorities, in close cooperation with the engineering teams.
Four questions show what that means in practice:
- What will the machine’s core technology building block be in five years?
- Should software development be brought in-house, or should the company rely on third-party components?
- Which architecture will guarantee the scalability of the product range at the lowest cost?
- What share of margin will come tomorrow from hardware, software or connected services?
These are executive-level choices. A CEO who understands them is far less likely to treat an engineering budget as a simple line to trim when markets weaken.
What happens to leadership teams that do not adapt?
In a sector where the machine is becoming software and obsolescence is accelerating, technology can no longer be relegated to the status of a support function. Leadership teams that continue to treat it as such will not lose a battle. They will drop out of the game.
Connect Bruno Bouygues on LinkedIn
Visit GYS to get more information
Also Read:-
How to Get Your First Startup Customer: A Simple Guide
Best Businesses to Start With $1,000
How to Build a One-Person Business Using AI Tools
