Next Shift

IMTS 2026: What Will Actually Change on Your Floor, and What Won't

In a week and a half, 90,000 people are going to walk into McCormick Place and be told that manufacturing has changed again. Some of them will come home with a signed quote. Most will come home with a tote bag, sore feet, and a vague, expensive feeling that their shop is behind. That second experience is the one worth talking about, because it happens every two years and it is almost entirely avoidable.

The problem is not that IMTS lies. Everything on that floor is real and most of it works. The problem is that a trade show is built to sell the front edge of the technology curve, and a 15-person job shop lives about five years behind that edge, by choice and by arithmetic. Nobody at the show is going to tell you which half of what you are seeing applies to you. So here is a translation: what a small shop can realistically act on out of IMTS 2026, what it cannot yet, and why walking the floor is still worth the plane ticket even if you buy nothing.

What the IMTS 2026 Show Floor Is Built to Sell

Start with the scale, because the scale is the point. IMTS 2026 spans four buildings at McCormick Place and more than 1.2 million square feet, with over 2,000 exhibitors and better than 3,000 conference registrants. The South Building holds metal removal and additive, the latter accelerated by Formnext. The West Building is tooling and workholding. The North Building carries automation, accelerated by SPS Smart Production Solutions, alongside abrasive machining, sawing and finishing, and gear generation, with the Smartforce Student Summit on Level 1 aimed at the skills gap. The East Building holds software, quality assurance, fabricating and lasers, and machine components.

Read that layout as a map of where the industry's money is going and it is accurate. Read it as a shopping list and it will hurt you. A floor that size exists because the exhibitors are selling to plants, not shops, and to five-year capital plans, not to the guy deciding whether he can justify a second machine this year. When a builder demonstrates an unattended cell with a robot, a pallet pool, in-process gauging and a dashboard, that demonstration is honest. It is also a system that costs more than your building. The gap between the demo and your floor is not credibility, it is capital, headcount and infrastructure, and nobody on the floor is incentivized to name it.

What Will Actually Change on Your Floor

The technologies that reach small shops share one trait: they arrive as an option on something you were buying anyway, or as a retrofit to something you already own. They do not require you to change how the shop works. Four categories qualify this cycle.

Probing and in-process measurement. A spindle probe and a tool setter are among the cheapest options on a machining center that actually change how a shop works, and they keep getting cheaper. They cut setup time, they catch a crash before it becomes scrap, and on the right job they move inspection off the surface plate and into the cycle. This is a four-figure decision on a new machine and a retrofit on an old one.

Bar feeders and simple pallet automation. The cheapest unattended hours available to a small shop are still a bar feeder on a lathe and a two-pallet changer on a mill. Neither is new, both are on the floor in force, and both come to the used market constantly. If you want spindle hours without hiring, start here rather than with a robot.

Small robot loading cells. The economics at the small end have moved, mostly because compact and collaborative arms now cost less than the machine they load. A compact loading cell on a single machine, running a family of parts, is now a decision a 15-person shop can pencil out, particularly used or as a package with a machine. The cell that does not pencil out is the one that requires re-engineering your workholding, your part presentation and your scheduling all at once.

Adaptive and control-side intelligence that ships standard. Load monitoring, chatter suppression, adaptive feed control and thermal compensation increasingly come with the control rather than as a project. You do not implement these. You buy a machine that has them and turn them on. That is the definition of technology that actually reaches a small shop.

The technology that changes a small shop is never the technology on the keynote stage. It is the option box on the quote, the retrofit kit, and the feature that came free on the control. If adopting it requires a project plan, it is not going to happen in a 15-person shop this year.

What Will Not, and the Honest Reason Why

None of what follows is fake. All of it works. It just does not reach a shop your size in this cycle, and it is worth knowing why so you can recognize the moment it does.

Humanoid robots. Real technology, early market. The barrier is not the robot, it is that general-purpose dexterity is the most expensive way to solve a problem a fixture and a two-axis loader already solve for a fraction of the cost. This becomes interesting to you when the labor it replaces is varied enough that dedicated automation cannot cover it, and that is not most small shops yet.

Full digital twins and plant simulation. The technology is real and the value is real at scale. The barrier is data. A twin is only as good as the model behind it, and building that model is a staffed, ongoing job. A shop without somebody whose actual role is manufacturing engineering will build a twin that is wrong within a quarter.

Machine learning on production data. Same barrier, different shape. Predictive models need history, and history needs collection infrastructure that has been running for a while. If you are not already logging machine data today, the value of this is at minimum two years out for you, and it starts with the boring step of collecting anything at all.

Additive as production. Additive is a large and growing part of the show, accelerated by Formnext in the South Building, and for tooling inserts, conformal cooling, fixtures and prototypes it is already useful. As a way to make your production parts, the part cost and the post-processing still favor cutting metal for almost everything a job shop quotes. Watch it, do not buy it, unless you have found the specific part that only works that way.

The Real Reason to Walk the Floor

Here is the part most pre-show coverage will not tell you. The machine being introduced at IMTS 2026 is the machine that starts hitting the used market in the early 2030s. That is not a knock on either the machine or the market, it is just how capital equipment moves. Most metalworking machinery depreciates over a seven-year recovery period in the United States, and in our experience buying and selling since 2008, a machine bought new turns over with its first owner somewhere between five and eight years in. So the floor in September is a fairly reliable catalog of what a small shop will be able to buy at a price that makes sense, later in the decade.

That reframes the trip completely. You are not shopping. You are learning which platforms, which controls and which option packages are going to be common on the used market in the back half of this decade, so that when one shows up you already know whether it fits your work and what to check. Walk it that way and the questions change. Not "what does this cost," but "which options on this machine will still matter in five years, which control generation is this, and how long will the builder support it." Those are used-buyer questions asked at a new-machine show, and they are the highest-value thing you can do at IMTS with no budget at all. And if the show does convince you to sign for something new, remember which side of that cycle you just moved to. The machine it replaces is the front end of somebody else's five-year plan.

Resell CNC Take

We see the back half of this cycle every day. The options a shop wishes it had bought, the control generations that aged well and the ones that stranded people, the automation that got installed and the automation that is still sitting on the pallet it arrived on. Go to the show. Look at everything. Then come back and buy the version of it that somebody else already paid the depreciation on, with the options they specified and you now get for free.

Frequently Asked Questions

What should a small shop actually expect from IMTS 2026?

Expect most of the floor to be aimed at plants with capital committees rather than at a 15-person shop. The technologies that will realistically reach a small shop within eighteen months are probing and in-process measurement, bar feeders and simple pallet automation, small single-machine robot loading cells, and adaptive control features that now ship standard. The ones that will not are humanoid robots, full digital twins, machine learning on production data, and additive as a production process.

When and where is IMTS 2026?

IMTS 2026 runs September 14 to 19, 2026 at McCormick Place in Chicago. The show spans the East, West, North and South buildings with more than 1.2 million square feet of exhibit space, over 2,000 exhibitors, roughly 90,000 expected visitors and more than 3,000 conference registrants. The 2026 theme is "Achieve the Impossible."

Is IMTS worth attending if I am not buying a machine?

Yes, and arguably it is more useful that way. The machines introduced at a show like this reach the used market roughly five to eight years later, so walking the floor is a preview of what a small shop will be able to afford in the early 2030s. Going with no budget removes the pressure and lets you ask better questions about control generations, option packages and long-term support.

What technology at IMTS can a 15-person shop actually afford?

The things that arrive as options or retrofits rather than as systems. A spindle probe and tool setter. A bar feeder on a lathe. A two-pallet changer on a mill. A compact robot loading cell on a single machine running a part family. Control-side features like load monitoring, adaptive feed and thermal compensation, which increasingly come standard rather than as a project. The common test is whether adopting it requires changing how the shop works.

What should I ignore at IMTS 2026?

Nothing, if you are looking to learn. But do not buy anything whose value depends on infrastructure you do not have. Digital twins and machine learning both need data collection and someone to own the model. Humanoid robots solve a labor problem that a fixture and a simple loader solve more cheaply in most small shops today. Additive is worth understanding for tooling, fixtures and conformal cooling, and is still hard to justify as a way to make production parts a job shop would otherwise cut.

How does a trade show announcement end up on the used market?

Capital equipment turns over on a fairly predictable cycle. Most metalworking machinery depreciates over a seven-year recovery period in the United States, and a machine bought new after a show typically turns over with its first owner somewhere between five and eight years in, when the shop replaces it, upgrades, or the business changes. That machine then enters the used market with its original option package intact, which is why a shop shopping used in the early 2030s will be looking at the iron introduced in 2026, often better equipped than it would have specified for itself.