Your Machine Shop’s Biggest Bottleneck Isn’t the Machines — It’s Everything Before Them
I’ve run enough Kaizen events on machine shop floors to know the machine is rarely the problem. This is the first in a four-part series on the bottlenecks I keep running into at aerospace and defense machine shops — planning the work, quoting it fast enough to win it, running customer service without flying blind, and being ready to run the job the second it lands on the floor. We’ll get to the other three in the posts that follow. This one’s about planning, and it starts with a number that still gets a reaction when I say it out loud: one Kaizen event, one process, $1 million a year.
The Problem: A 13-Person Handoff Chain Before a Part Ever Reached the Floor
The client is a growing aerospace and defense contract machinist — precision, make-to-print work, the kind that ends up on landing gear and other flight-critical components. Nothing about the parts themselves was the problem. These guys can run a machine. What they couldn’t do was move a customer’s print and spec package through their own system without burning an absurd amount of labor on it.
Here’s what that looked like before we touched it: a customer sends a print, a material spec, and a pile of other requirements — sometimes dozens of pages. To turn that into something the shop floor can actually run, someone has to enter all of it into the system. At this client, that “someone” was actually thirteen or fourteen people. The work got handed from group to group in sequence, and — this is the part that should bother you — every single group had its own checker standing behind it, re-verifying what the last group did.
When we started mapping that process on the wall, the room got quiet. Nobody had ever looked at it end to end before. And when we asked the obvious question — of everything happening in this chain, how much of it is actually value-added? — the honest answer was two or three steps. Out of the whole thing.
That’s the real lesson here, and it’s bigger than this one client: if your quality strategy depends on someone double-checking someone else’s work, you don’t have a quality process. You have a cost center built to catch your own mistakes. In lean, we say it plainly — do it right the first time. Don’t use inspection as a crutch. This team had built an entire department on the crutch.
Getting Past the Fear: How Leadership Held the Line
That reaction showed up in the room almost as soon as the map went up on the wall — and working through it is half of what a Kaizen event like this is actually for. Once the team saw their own process laid out in front of them, they got nervous. Fast. Because the honest version of “we’re only doing value-added work” means fewer people doing the same job, and everybody in that room could do that math themselves. This is exactly the point where most improvement efforts quietly die: leadership gets uncomfortable with what the map is telling them and lets people add the old steps back in “just to be safe.” Our job at that moment is to keep the team looking at the data long enough for leadership to actually define the future state, instead of retreating to the current one. Here, management held the line: we are not doing non-value-added work, full stop. That conviction — not any tool or template we brought in — is what made the redesign stick.
The Fix: One Person Owns the Process, Start to Finish
The redesign itself is almost anticlimactic once you’ve done the hard part. Instead of relaying a spec through a chain of groups and checkers, one person now owns the entire intake task, start to finish. No handoff, no re-check, no crutch. Modeled all the way out, that department can run at roughly half its original headcount. We didn’t flip a switch — the rollout was phased, with two people running the new process immediately, covering close to half the workload, then scaling up from there.
The Result: $1 Million a Year in Labor Productivity
The number that came out of it: about $1 million a year in labor productivity. And to be clear about what that means for the people involved — this wasn’t a layoff. This is a company growing from around $100 million toward $130 million in revenue. The labor that came off this process didn’t walk out the door; it got redeployed into work that actually adds value, and normal attrition took care of the rest. You don’t need to choose between efficiency and your people. You need to stop wasting both on work that was never adding value in the first place.
What’s Next: Using AI to Replace the Human Checker
One more thing worth flagging, because it’s where this is headed: aerospace is understandably cautious about AI, so this client isn’t there yet. But the obvious next step is letting AI check completed work against spec instead of routing everything through a human checker. That’s not theoretical — it’s the same problem we just solved, one layer further down.
Is Your Intake Process Hiding the Same Waste?
Here’s the uncomfortable question I’d ask if I were running your shop: how many “just to double-check” steps are quietly built into your own intake process, and what are they costing you every year? If you’re a contract manufacturer for aerospace or defense, I’d bet money you’re running some version of this same chain right now. Most shops are.
If you want a second set of eyes on your own process, TBM does exactly the kind of value-stream mapping and Kaizen work that found this $1 million — happy to talk through what that would look like for your operation.
Next in this series: how the same shop cut two weeks out of its quoting process — and what that meant for revenue, not just labor.
