How to Order Custom Manufacturing Without Getting Burned: A 7-Step Checklist

In my role coordinating custom manufacturing and prototyping, I've handled 200+ rush orders in the last four years. In Q1 2025 alone, we processed 47 rush orders with 95% on-time delivery. That stat didn't happen by accident. It happened because I stopped trusting verbal promises and started using a checklist.

This checklist is for anyone ordering 3D printed parts, CNC machined components, laser cut pieces, or injection molded runs. It's also for people just starting out who aren't sure what to ask. Not a deep dive into additive manufacturing theory. A practical set of checks that takes about ten minutes to run.

Here's the full list: start with the part requirement, verify the process, confirm tolerances, plan post-processing, lock the lead time, compare total cost, and test a small batch. Seven steps. Do them in order.

Step 1: Define the part requirement, not the technology

Start with function. What does this part actually have to do? Is it a cosmetic prototype, a production jig, or a bracket that holds 40 pounds? The answer changes everything.

Most people ask whether they should use 3D printing or CNC. The better question is what stresses the part will see and how many units you need. I had a client in March 2024 call at 4:30 PM with a display booth opening at 9 AM the next day. They needed 12 machined brackets. Normal lead time was five business days. We got them made by 6:50 AM because we already knew the exact load, material, and surface requirements. No time for back-and-forth.

Checkpoint: Write down the part's purpose, environment, max load, operating temperature, and quantity before contacting anyone.

Step 2: Verify the process and machine, not just the brand name

Brand names matter less than the specific machine and the person running it. I've seen great parts come off budget equipment and bad parts come off expensive equipment. The difference is usually setup and inspection.

When I research a supplier, I don't just read 3D-Systems metal 3D printer reviews and move on. A review tells me what that machine can do in ideal conditions. It doesn't tell me whether the shop calibrates it daily or uses worn materials.

Ask for the exact machine model. Ask how long the operator has run it. Ask for examples of parts similar to yours. If they can't show you one, that's a signal.

Also, if a supplier tries to impress you with advanced systems you don't need, be careful. 3D laser projection systems are genuinely important in aerospace for composite layup and assembly alignment. They're not what you need for a simple prototype bracket. A good supplier educates you instead of confusing you. I'd rather spend 10 minutes explaining options than deal with mismatched expectations later.

Checkpoint: Get the machine model, the operator's experience level, and at least two relevant example parts.

Step 3: Confirm tolerances and surface finish in writing

This is where most disagreements start. CAD looks perfect. Reality has layer lines, tool marks, and thermal shrinkage.

Ask what tolerance applies to critical dimensions. For CNC parts, that might be ±0.005 inches. For 3D printed parts, find out whether the tolerance applies to the printed part or the final machined surfaces.

People also ask me what are good 3D printers for beginners. My short answer: a printer with good community support, cheap materials, and a simple workflow. But that's for learning. For production parts, you're not buying a printer. You're buying a documented process.

Get the surface finish defined too. A raw FDM part is not equivalent to a vapor-smoothed part, even if they use the same material.

Checkpoint: Require a drawing or spec sheet with tolerances, not just a verbal 'yeah, it'll be fine.'

Step 4: Plan post-processing before you approve the quote

Most people ignore this step. Actually, that's not quite right—they ignore it until the parts arrive, and then they're disappointed.

Do you need holes tapped? Edges broken? Supports removed? Parts bead blasted, anodized, or painted? Each one adds cost and time.

Let me use a weird example: need a 2 inch pvc cutting tool? You don't just need something that cuts pipe. You need a blade with the right arbor size, the right tooth geometry, and a guard that won't get in the way. The extras are the product.

Custom manufacturing is the same. A part straight off the machine is not a finished part until you define what finished means.

Checkpoint: Ask 'What does this price include for finishing?' If the answer is 'standard,' ask them to define it.

Step 5: Lock lead time in writing and add your own buffer

Here's the thing: suppliers quote optimistic lead times. They want the order. Then reality happens.

If a vendor says five days, I plan for eight. If that buffer feels unnecessary, great. If it saves you once, it's worth it.

Why do rush fees exist? Because unpredictable demand is expensive to accommodate. I have mixed feelings about them. On one hand, some feel like gouging. On the other, after seeing the operational chaos a late order causes, maybe they're justified. Either way, avoid needing them.

We created a policy the hard way in 2023. We lost a $12,000 contract because we tried to save $80 on standard shipping. When the standard delivery missed the deadline, the rush replacement cost $400 (which, honestly, stung more than the $80). Now we build a 48-hour buffer into every client commitment.

Checkpoint: Get the promised ship date in the quote. Then add 20-30% buffer before sharing the date with your client.

Step 6: Compare total cost, not price per part

The quote says $4.00 per part. At 500 units, that's $2,000. But what about setup, tooling, minimum order quantities, packaging, and freight? A cheap part can become an expensive order.

Ask for an itemized quote. If they won't itemize, ask why. Some shops bundle pricing, and that's fine (as long as you know what's inside the bundle).

In January 2025, I requested quotes from four suppliers for identical machined parts. Prices varied by almost 40%. The most expensive was not the best. The cheapest was not the worst. Total cost, lead time, and communication mattered more. Prices as of January 2025; verify current rates.

Checkpoint: Compare total landed cost: setup + unit price + finishing + shipping + potential rush fees.

Step 7: Test with a small batch before scaling

Never skip this step. I know you're in a hurry. That's exactly when skipping it hurts the most.

Samples are often made with extra care. Production runs are made at speed. So run a trial batch of 5-10 parts first, especially with a new supplier.

We didn't have a formal first-article inspection process for a long time. The third time we ordered the wrong quantity, I finally created a verification checklist. Should have done it after the first time.

Checkpoint: Require a first-article inspection report. Check critical dimensions, surface finish, and material certs before approving full production.

Mistakes I've Made So You Don't Have To

A few reminders from the messy side of this industry:

• Skipped the buffer. Saved $80 on shipping, spent $400 on a rush replacement. See Step 5.
• Trusted the 'we do it all' supplier. Some really do. Most are great at one process and average at the rest. Ask what they run daily, not what they can technically do.
• Assumed every part was inspected. Unless you ask, 'we check everything' is a hope, not a process.
• Confused brand reputation with current performance. A known brand can have a bad batch. A small shop can be outstanding. Verify the actual parts.

One more thing: if you're hiring for additive manufacturing jobs UK employers care about hands-on troubleshooting more than certificates. A candidate who can explain why a build failed is worth more than one who only knows how to load a file.

This checklist won't make a five-day lead time disappear. It won't make a bad part good. But it will help you ask the right questions before you spend money. That's the whole point.

Look, I'm not saying budget options are always bad. I'm saying risk should be intentional. If you need parts in a hurry, a checklist is what keeps you from making the same mistake twice.