Who This Checklist Is For
You're a procurement engineer at a defense contractor or aerospace company. You have a complex part—think internal cooling channels, lattice structures, or geometries that make a CNC machinist swear—and you need it yesterday. Maybe it's a prototype for a government review, a critical replacement for a test stand, or a last-minute design change on a production run.
This checklist is for you. It's not a deep dive into the physics of metal additive manufacturing (AM) vs. subtractive CNC. It's a practical, step-by-step decision framework I've developed over coordinating about 200 rush orders for high-stakes projects, including several for clients who eventually became Velo3D customers. It covers the five questions you need to answer before you pick up the phone—and why the obvious answer isn't always the right one.
Here are the five steps:
Step 1: Assess Your Urgency—and Your Honesty About It
Everyone says their part is urgent. But there's urgent—and then there's urgent.
In March 2024, I had a client whose vendor had ghosted them on a titanium bracket for a military drone prototype. They called at 2 PM on a Friday, needing the part by Monday morning for a live-fire test. That's urgent. Their normal CNC shop couldn't touch it in under two weeks. We had to get creative.
Here's how I trip age a rush order now. Ask yourself three things:
- What's the real deadline? Not the 'I'd like it by…' but the hard stop. A contract penalty, a test window, a customer ship date. Write it down.
- What's the cost of missing it? A $50,000 penalty clause? Losing a government contract worth millions? Or just some angry emails? The higher the cost, the more you should pay for certainty.
- Can the part be simplified? This is the one most buyers miss. They assume the design is locked. But sometimes a 24-hour redesign can remove a feature that adds two weeks to the lead time. I've seen a machined bracket become a weldment in an afternoon, cutting delivery from three weeks to four days.
The surprise here isn't complexity—it's that many buyers don't even ask question #3. They focus on getting the exact spec made, and they completely miss the option of a slightly different spec that can be delivered five times faster. My experience is based on mid-to-high-complexity aerospace parts. If you're working with simple geometries, your mileage may differ.
Step 2: Know Your Part's Geometry—Is AM Actually Faster?
Conventional wisdom says CNC is faster for simple parts, and AM is faster for complex ones. That's true—but it's too blunt an instrument. The real question is: what's the total path to a finished part?
For CNC, the path includes: programming the CAM, fixturing, multiple set ups for different features, tool changes, and secondary operations. For a simple block with holes, that's a day. For a complex bracket with undercuts, that's a week—or more, if you need special tooling.
For AM (specifically powder bed fusion like Velo3D's process), the path includes: file preparation and support generation (though Velo3D's approach is great for reducing supports), printing (which can be 12-72 hours depending on size and height), and post-processing (heat treat, support removal, surface finishing). The print time is fixed; it doesn't care about internal complexity. That's the killer advantage for urgent parts.
Here's the decision rule I use:
- CNC wins for: Parts with simple geometries, standard materials (6061 aluminum, 304 stainless), and no complex internal features. If you can make it on a 3-axis mill in two days, stick with CNC.
- AM wins for: Parts with internal channels, lattice structures, organic shapes, or where you want to consolidate multiple parts into one assembly. The more complex, the more AM pulls ahead—especially when you factor out the time to make fixtures and tooling.
But here's the blind spot most buyers have: they think the comparison is just about print time vs. machine time. It's not. It's about the total time to deliver a usable part. For a CNC part, that includes waiting for the machine to be free, tooling procurement, setup, and inspection. For an AM part, it includes the build, then post-processing, which can take as long as the print.
Step 3: Determine What 'Velo3D' Actually Buys You
Not all metal AM is the same. Velo3D's claim to fame—and it's a meaningful one for urgent parts—is that their process can handle complex geometries without requiring sacrificial support structures in many cases. Support structures are a huge pain: they add print time, require manual removal, and can limit design freedom. Reducing or eliminating them is a real advantage when time matters.
Why does this matter? Because if your part has overhangs that need support in a standard AM system, you're looking at extra print time, plus machining time to cut off the supports, plus risk of damaging the part. With Velo3D's approach, you can often skip that step entirely. For a part I saw in Q3 2024, a client needed an impeller with internal vanes. A competitor quoted 48 hours of print time plus 16 hours of post-processing. Velo3D's system printed it in 38 hours with minimal support removal.
That's not to say Velo3D is the answer for every urgent AM part. It's a higher-cost system, and the machines are less widely available than, say, a standard EOS or SLM system. If you don't have access to a Velo3D machine in your supply chain, you can't use it.
Step 4: Build a Vendor Triage Protocol
You cannot wait for quote cycles when a deadline is looming. Here's the three-category system I've developed:
Tier 1: Proven Rush Vendors (Velo3D Certified, if available)
- Pre-approved for urgent work
- Pre negotiated rush fees (20-50% premium, depending on complexity)
- They have your material specs and design files on file
- You can call them and say 'I need this in X hours, can you do it?'
Tier 2: Reliable Standard Vendors (CNC shops you've used)
- They can handle simple parts fast
- But don't expect miracles on complex parts
- Have a backup vendor in this tier
Tier 3: Emergency Options (Local shops, NRE-heavy solutions)
- These are your 'break glass in case of emergency' options
- Expect to pay 2-3x normal rates
- Expect non-optimal quality control
The question everyone asks is 'what's your best price?' The question they should ask is 'do I have a vendor who can deliver this in 48 hours if I need them to?'
Step 5: Verify Before You Commit
This is where most mistakes happen. You get a quote for a 36-hour turnaround, you say 'yes,' and then you find out the vendor's definition of 'delivery' doesn't include post-processing or shipping.
Here's what to verify:
- Print time vs. total time. 'Available in 24 hours' likely means the print finishes in 24 hours, but you still need to remove supports, do surface finishing, and inspect. Ask: 'What time on [date] will the part be ready to ship?'
- Shipping time. If the vendor is in California and you're in Virginia, 'parts ready Wednesday' means nothing if the shipping takes three days.
- Material certs. For defense and aerospace, you likely need a material certification with the batch number. If the vendor doesn't have the right stock, your 36-hour turnaround turns into a week waiting for material.
- Inspection. Who does the NDT? Is it included? A part that fails inspection on Friday means your Monday deadline is gone.
The surprise isn't usually the price. It's the assumption that the vendor is solving for the same constraints you are. I've only worked with U.S.-based vendors for these high-stakes parts. If you're sourcing internationally, these principles apply, but the logistics are a different beast.
Final Thought: The Cost of 'Probably'
I'll leave you with this. In 2023, our company lost a $250,000 contract because we tried to save $1,200 on a rush order. We went with a CNC shop that said 'probably' they could deliver on time. They didn't. The client went with a competitor who had the part in hand a day earlier, from an AM vendor.
The value of guaranteed delivery isn't the speed. It's the certainty. When you're looking at a 36-hour turnaround and a $50,000 penalty for missing it, the vendor who says 'yes' and means it is worth more than the one who says 'probably' and is half the price. Budget for the premium. Never bet on 'probably.'
This was true five years ago when the AM industry was less mature. It's even more true today, when the technology is proven but the supply chain is still catching up. The tools are there—Velo3D's machines, certified vendors, simulation software. The hard part is making the right call in an hour, not a week.