Inside Kramer Industries: The Machines That Changed Precision Machining

I’ve been inside over 400 machine shops across 40 countries, and I can usually tell within five minutes whether a shop is running well or just surviving. It’s not about the machines — every shop has decent iron these days. It’s about the flow. How material moves from receiving to shipping. Where the bottlenecks are. Whether the guy running the Mazak actually understands what the program is doing, or if he’s just pressing the green button.

Kramer Industries in Grand Rapids, Michigan, is one of those shops where you walk in and immediately know something is different.

The shop that almost didn’t make it

Dave Kramer — third generation — will tell you straight up that 2019 almost killed the business. They were running 14 machines: a mix of old Haas VF-2s, a couple of Mori Seikis from the early 2000s, and three manual Bridgeports that his grandfather bought in 1978. Twenty-three employees. Revenue had been flat for six years.

“We were a general job shop,” Dave told me as we walked past the old manual section, which is now a break room. “We’d quote anything. Aluminum housings, steel brackets, the odd one-off prototype. We were competing on price against shops in Mexico and China, and we were losing.”

The turning point came when their biggest customer — a Tier 2 automotive supplier — told them they had 18 months to achieve AS9100 certification or lose the contract. That contract was 40% of their revenue.

“I sat in my truck in the parking lot after that meeting and seriously considered just selling the building,” Dave said. “My wife asked me what I wanted to do, and I said, ‘I want to build the shop my grandfather would have built if he had the technology.’”

The $2.8 million bet

What followed was the most aggressive capital investment I’ve seen from a shop this size. Dave secured an SBA loan and committed $2.8 million over 24 months. The shopping list:

Two DMG MORI DMU 50 5-axis machines with pallet systems
One Mazak INTEGREX i-200 multi-tasking center
A Zeiss CONTURA CMM with CALYPSO software
Full Mastercam 2024 with 5-axis simultaneous capability
A Fastems FPC pallet automation system connecting both DMU 50s

That last item is what changed everything.

I stood in front of the Fastems cell for a good twenty minutes, watching it cycle. The system holds 24 pallets, each fixtured with a different part. The two DMU 50s pull pallets from the system, machine the parts, and return them — all without human intervention. Dave runs this cell lights-out on second shift, producing parts for 16 hours with nobody in the building.

“The Fastems paid for itself in nine months,” Dave said. “Not because it made us faster — although it did. Because it eliminated the setup time that was killing us. We were spending 30% of our spindle time on setups. Now that number is under 8%.”

The floor tells the story

Here’s what I noticed walking the floor:

Material flow is linear. Raw stock comes in the east door, moves through sawing, then to the 5-axis cell or the Mazak, then to inspection, then to shipping on the west side. No backtracking. No carts of WIP sitting in corners waiting for someone to remember they exist.

Every machine has a purpose. The two old Haas VF-2s are still running — but they handle all the 3-axis work that doesn’t justify tying up the 5-axis machines. Simple brackets, fixtures, soft jaws. They’re not prestige machines, but they’re earning their keep.

The CMM isn’t a bottleneck. This is where a lot of shops get it wrong. They invest millions in cutting machines and then inspect parts with calipers and a granite plate. Dave put the Zeiss in before the 5-axis machines arrived. His logic: “You can’t produce parts to aerospace tolerances if you can’t measure them. The CMM had to come first.”

Tooling is organized to a level I rarely see. Every tool in the crib has a barcode. When a tool is pulled, it’s scanned, assigned to a job, and tracked through its life cycle. Worn tools are flagged automatically. This wasn’t a $500,000 tool management system — Dave’s lead programmer built it with a barcode scanner, a Raspberry Pi, and a Google Sheet. Sometimes the best solutions are the scrappy ones.

The people problem (that isn’t what you think)

Dave went from 23 employees to 14. But here’s the thing everyone gets wrong about automation: he didn’t lay anyone off. Nine people left through natural attrition — retirements, people who moved, one guy who went back to school. The 14 remaining employees are better paid, better trained, and producing 40% more revenue per person than before.

“I used to hire bodies,” Dave said. “Now I hire thinkers. My youngest machinist is 24 and he’s programming the 5-axis cell. He came in with some community college CAM training and we developed him. He’s worth more to me than three button-pushers.”

This is the narrative that gets lost in the “automation kills jobs” discourse. In every shop I tour that has successfully automated, the remaining workforce is more skilled, more engaged, and better compensated. The jobs that disappeared were the ones nobody wanted in the first place — standing in front of a machine for eight hours loading blanks.

The numbers that matter

Dave shared his metrics with me (he’s unusually open for a shop owner, which I respect):

Revenue per employee: $285,000 (up from $168,000 pre-automation)
On-time delivery: 97.2% (was 81%)
Scrap rate: 0.8% (was 4.1%)
Setup time as % of spindle time: 7.6% (was 31%)
Lights-out hours per week: 64

The AS9100 certification? They got it in 11 months. The automotive customer that gave them the ultimatum now sends them more work than they can handle. Dave is shopping for a third 5-axis machine.

What I’d steal from this shop

Every factory tour, I identify three things I’d take back to my own operation. From Kramer Industries:

CMM first, cutting machines second. If you’re investing in tighter-tolerance work, buy the measurement capability before the machining capability. You can’t improve what you can’t measure.
Pallet automation is the real game-changer. Everyone obsesses over spindle speed and rapid rates. The real competitive advantage in 2026 is setup reduction. The Fastems system didn’t make Dave’s machines cut faster — it made sure they’re always cutting.
Track your tooling like you track your jobs. Dave’s DIY tool tracking system cost him $800 and saves him thousands per month in emergency tool orders and scrapped parts from worn inserts. If a third-generation shop in Michigan can do this with a Raspberry Pi, you have no excuse.

The walk back to the truck

I always take a minute in the parking lot after a factory tour to process what I saw. Kramer Industries isn’t the biggest shop I’ve visited. It’s not the most technologically advanced. But it might be the best example I’ve seen of a small shop doing exactly the right things at exactly the right time.

Dave Kramer bet his family’s legacy on a technology upgrade. He didn’t just buy machines — he reimagined how work flows through his building. He invested in measurement before machining. He chose automation that eliminates waste, not just labor.

And he did it all while taking care of his people.

That’s precision manufacturing in 2026. The shops that will survive the next decade aren’t the ones with the most machines. They’re the ones with the clearest thinking.

Tony Gunn has visited 500+ manufacturing facilities across 70+ countries. He is the host of The Machinists Club podcast and CEO of TGM Global.