R&D Tax Credits for Custom Manufacturers: What Counts, What Doesn't

TL;DR

• Custom and contract manufacturers frequently solve technical problems on the shop floor (new tooling, process changes, material substitutions) that may qualify for the federal R&D credit under IRC Section 41.
• Qualifying work has to clear the IRS four-part test. Routine production runs, cosmetic changes, and simple quoting usually do not count.
• A no-cost assessment can help you separate qualifying activity from routine work before you commit to anything.

When a customer hands you a drawing and says "can you make this?", what happens next often looks a lot like research and development, even if nobody in your shop calls it that. You may be designing new fixtures, testing tolerances, trialing materials, or reworking a process until the part comes out right. That problem-solving can sometimes qualify for the federal research and development (R&D) tax credit, and many custom manufacturers leave it on the table because they assume the credit is only for people in lab coats.

This post walks through what the federal R&D credit covers, what qualifying activity can look like in a custom or contract manufacturing shop, and the kind of work that typically does not qualify. The goal is to help you tell the difference before you spend time chasing it.

What the federal R&D credit actually covers

The federal credit comes from Section 41 of the Internal Revenue Code, with the qualifying-activity rules spelled out in Treasury Regulation 1.41-4 (ecfr.gov). It is meant to reward businesses for the work of developing or improving products and processes, not just for inventing something brand new to the world. Improving how your shop makes a part can count, even if the underlying idea is not novel to the industry.

To qualify, an activity generally has to pass the IRS four-part test. All four parts have to be met:

First, the work must be technological in nature, meaning it relies on principles of engineering, physics, chemistry, computer science, or a similar hard science. For a machine shop, that usually means mechanical or materials engineering rather than aesthetics or guesswork.

Second, it must serve a permitted purpose: developing a new or improved product or process, where the improvement relates to function, performance, reliability, or quality.

Third, there must be technical uncertainty at the outset. You did not already know whether you could achieve the result, or how, or what design or method would work.

Fourth, the work must involve a process of experimentation, meaning you evaluated alternatives through modeling, simulation, trial and error, or systematic testing to resolve that uncertainty.

If an activity clears all four, it may qualify. If it misses even one, it generally does not.

What qualifying activity can look like in a custom manufacturing shop

In custom and contract manufacturing, qualifying work often hides inside everyday jobs. Depending on the specific activities, documentation, and facts, the following may qualify:

Designing and testing new tooling, jigs, or fixtures to hold or machine a part that your existing setups cannot handle. Developing a new manufacturing process, or significantly reworking an existing one, to hit a tolerance, finish, or cycle time you could not previously achieve. Trialing a different material or alloy when a customer's spec or a supply change forces you to figure out how it behaves under your processes. Programming and iterating CNC toolpaths to solve a chatter, deflection, or heat problem on a difficult geometry. Building and refining a prototype or first article through multiple iterations before a part is production-ready. Engineering automation or a new fixturing approach to make a previously manual or unrepeatable process reliable.

The common thread is uncertainty plus experimentation. You were not sure it would work, and you had to test your way to the answer.

What usually does NOT qualify

It is just as important to be honest about what tends to fall outside the credit, because over-claiming is where shops get into trouble. The following generally do not qualify:

Routine production runs of a part you already know how to make. Cosmetic or stylistic changes that do not affect function or performance. Routine quality control, inspection, and testing of finished goods. Quoting, estimating, and other administrative or sales work. Simply reproducing an existing part from a complete customer drawing with no technical uncertainty on your end. Work where you already knew the method and outcome going in, even if the job was difficult or time-consuming. Difficulty alone is not the test; uncertainty resolved through experimentation is.

Routine duplication and straightforward production are the bulk of what most shops do, and that is fine. The credit is about the slice of work where you had to engineer a solution.

Why a careful read matters here

Custom manufacturing sits in a tricky spot. There is often real qualifying activity, but it is mixed in with a lot of routine production, and the line between the two depends on the facts. An overly conservative read can leave legitimate credit unclaimed. An overly aggressive read can pull in routine work that does not belong and raise audit risk. The aim is to identify and document the activity that genuinely meets the four-part test, and to leave out the work that does not.

FAQ

We make parts to customer drawings. Can we still qualify?
Possibly. It depends on where the technical uncertainty lives. If a customer hands you a complete, fully specified drawing and you simply reproduce it with known methods, that work usually does not qualify. If you have to develop the tooling, process, or approach to actually make the part, and the path was not certain at the outset, that development work may qualify.

Does it matter that the credit is for our customer's product, not ours?
Not necessarily. The relevant question is whether your shop performed qualifying activity and bore the associated risk and rights, which is a facts-and-circumstances analysis. Contract manufacturers can qualify, but the terms of the customer relationship matter, so this is worth reviewing carefully.

What documentation do we need?
Generally, records that connect the activity to the four-part test: design iterations, test results, scrap and rework tied to development, time spent by engineers and machinists on qualifying work, and the wages, supplies, and contract costs associated with it. Better contemporaneous records make a claim more defensible, though no documentation can guarantee a particular outcome.

How do we start without a big commitment?
Strata offers a no-cost initial assessment to gauge whether there is qualifying activity worth pursuing. A typical engagement runs about four to eight weeks depending on the complexity and the records involved. You can reach us at stratataxgroup.com/contact.

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