R&D Tax Credits for Custom Manufacturers: When Shop Floor Problem Solving May Qualify

TL;DR

• Custom manufacturers often assume that making parts to a customer's print means no R&D, but the credit rewards how you figure out production, not whether you invented the product.
• First articles, fixture and tooling design, process development, and scrap from trial runs can all contribute to a claim when the IRS four-part test is met and documented.
• Routine production, cosmetic changes, and problems solved by ordering a different tool from a catalog typically do not qualify, and an honest assessment separates the two.

"We Don't Do R&D. We Just Make Parts."

If you run a custom manufacturing shop, you have probably said some version of this. The customer sends a print, your team makes the part, and nobody in the building wears a lab coat. So the federal R&D tax credit must be for someone else, right?

Not necessarily. The credit was written for exactly the kind of problem solving that happens between "here's the print" and "here's the first good part." If your team has ever burned a week figuring out how to hold an odd geometry without distortion, chased tolerances through three trial runs, or built a custom fixture because nothing off the shelf would work, that work may be worth a closer look.

What the Federal R&D Credit Actually Covers

The credit is defined in IRC Section 41 and Treasury Regulation 1.41-4. To qualify, an activity generally needs to pass the IRS four-part test:

1. Technological in nature. The work relies on principles of engineering, physical sciences, or computer science. Machining, welding metallurgy, forming, and CNC programming all rest on engineering principles.
2. Permitted purpose. The work aims to create or improve a product or process in terms of function, performance, reliability, or quality. Improving your process to hit a spec counts, even when the customer owns the product design.
3. Technical uncertainty. At the outset, your team did not know whether it could achieve the result, or how, or what the right design would be. "Can we hold plus or minus half a thou on this thin-walled part without chatter" is a textbook example.
4. Process of experimentation. You evaluated alternatives through modeling, simulation, trial runs, or systematic testing rather than knowing the answer up front.

The often overlooked point for job shops: the credit can apply to process development, not just product invention. You do not need to own the design on the print to have qualifying activity in how you manufacture to it, depending on the specific activities, contract terms, and facts.

What Qualifying Activity Can Look Like in a Custom Shop

Depending on the facts and documentation, activities like these may qualify:

• First article development. The engineering and trial machining it takes to get from print to an approved first article, including CAM programming iterations, speed and feed development, and dimensional troubleshooting.
• Fixture and tooling design. Designing and testing custom workholding, jigs, or dies where the right approach was not known at the start.
• Process development and qualification. Developing weld procedures, heat treat recipes, coating processes, or machining sequences to meet a spec your standard process could not hit.
• Material and tolerance challenges. Experimenting with exotic alloys, thin walls, deep pockets, or stack-up issues where achieving the tolerance was genuinely uncertain.
• Scrap and trial runs. Material consumed in test runs during development can often be included as a supply cost when tied to a qualified activity.

Wages for the machinists, programmers, and engineers doing this work, plus a portion of supervision and direct support, are typically the largest component of a claim.

What Usually Does Not Qualify

An honest assessment matters as much as an ambitious one. Work that typically falls outside the credit includes:

• Routine production runs after the process is proven out
• Repeat jobs quoted and run the same way as before
• Cosmetic or stylistic changes with no technical uncertainty
• Problems solved simply by purchasing a different standard tool or machine
• Routine quality control inspection on production parts
• Administrative work, quoting, and general maintenance

If a claim treats every shop hour as R&D, that is a red flag, not a bigger credit. The defensible approach identifies the specific projects where real technical uncertainty existed and documents them.

What This Looks Like for an Owner or CFO

For a shop in roughly the $3M to $15M revenue range, the practical question is whether enough qualifying activity exists to justify a study. A few signals that it may:

• Your engineers or lead machinists spend meaningful time on new part numbers, prototypes, or process improvements rather than pure repeat work
• You regularly take on jobs other shops turned down because they were hard
• You have scrapped material figuring out how to make something work

The starting point does not need to be a commitment. Strata offers a $0 initial assessment to evaluate whether your activities may qualify, and a typical study runs about 4 to 8 weeks depending on complexity and how accessible your records are. Time records, job travelers, ECOs, first article reports, and scrap logs you already keep often form the backbone of the documentation.

FAQ

Our customers own the designs. Can we still qualify?
Possibly. When a customer owns the product design, the manufacturer's qualifying activity is often in process development. Contract terms matter here, particularly who bears the financial risk of the development work and who has rights to the resulting process knowledge, so this deserves a careful review of your agreements.

Do we need engineers on staff to claim the credit?
No engineering degrees are required. What matters is that the work relies on engineering or scientific principles. Experienced machinists and CNC programmers doing systematic process development can perform qualifying activity.

We already filed this year. Is it too late?
Generally, credits can be claimed on amended returns for prior open tax years, subject to specific IRS documentation requirements for refund claims. A tax professional can help you evaluate what is available in your situation.

Will claiming the credit trigger an audit?
No one can promise any filing outcome. What a well-prepared claim can do is hold up if questioned: credible qualification decisions, contemporaneous documentation, and consistent methodology. That is the standard we build to. You can review the IRS's own overview of the research credit on IRS.gov.

Next Step

If some of the projects described above sound like your shop floor, a conversation costs nothing. Request a $0 initial assessment and we will help you evaluate whether your work may qualify.

This post is for informational purposes only and does not constitute tax or legal advice. Consult a qualified tax professional regarding your specific circumstances.