Our last article focused on the 2025 updates to the form 6765 with regards to business components and the IRS’ increased focus on business component reporting. Let’s take a deeper look at the law, examples of qualified and non-qualified process of experimentation, highlight how the Tax Court is applying those rules, and illustrate best practices for documenting your R&D credit.
Qualified Research under the Internal Revenue Code
Under Treas. Reg. § 1.41-4 “Research constitutes qualified research only if it is research—
- With respect to which expenditures may be treated as expenses under section 174
- That is undertaken for the purpose of discovering information that is technological in nature, and the application of which is intended to be useful in the development of a new or improved business component of the taxpayer; and
(iii) Substantially all of the activities of which constitute elements of a Process of Experimentation that relates to a qualified purpose.”
The final requirement may raise a couple of questions, namely:
- How much is substantially all?
- What is a Process of Experimentation?
- The “substantially all” requirement is satisfied if 80 percent or more of a taxpayer’s research activities constitute elements of a process of experimentation that relates to a qualified purpose.
The percentage can be measured on cost or other consistently applied reasonable basis and is applied separately to each business component. If the business component doesn’t reach the minimum threshold, the company must apply the test at the next sub-level of the business component (shrink back rule).
- A Process of Experimentation includes:
- identifying technical uncertainty concerning the development or improvement of a business component with regard to capability, method or the appropriate design as of the beginning of the taxpayer’s research activities;
- identifying one or more alternatives intended to eliminate that uncertainty; and
- conducting a process of evaluating alternatives (through, for example, modeling, simulation, or a systematic trial and error methodology).
Process of Experimentation Examples
| Qualified | Non-Qualified |
| X seeks to improve fuel economy in its current model vehicle. X designs, models, simulates, tests, refines, and re-tests several alternative designs for the hood and associated proposed modifications to both the air intake system and cooling system. | Upgrade its warehouse management software. X evaluates several of the alternative warehouse management software products available from vendors in the marketplace to determine which product will best serve X’s technical requirements. X selects vendor V’s software. |
| X must develop load balancing software across a server cluster supporting multiple web applications. X is uncertain of the appropriate design of the load balancing algorithm, given that the existing evolutionary algorithms did not meet the demands of their highly volatile web environment. Therefore, X designs and systematically tests and evaluates several different algorithms that perform the load distribution functions. | X wants to develop software to balance the incoming processing requests across multiple web servers that run the same set of software applications. Without evaluating or testing any alternatives, X decides that a separate server will be used to distribute the workload across each of the web servers and that a round robin workload distribution algorithm is appropriate for its needs. |
| X determines that it must interface part of its legacy software with the new ERP software because the ERP software does not provide a particular function that X requires for its business. As a result, X must develop an interface between its legacy software and the ERP software, and X evaluates several data exchange software applications and chooses one of the available alternatives. X is uncertain as to how to keep the data synchronized between the legacy and ERP systems. Thus, X engages in systematic trial and error testing of several newly designed data caching algorithms to eliminate synchronization problems. | X wants to install an enterprise resource planning (ERP) system that runs off a single database. X evaluates its business needs and the technical requirements of the software, such as processing power, memory, storage, and network resources. X devotes the majority of its resources in implementing the ERP system to evaluating the available templates, reports, and other standard programs and choosing among these alternatives in configuring the system to match its business process and reengineering its business process to match the available alternatives in the ERP system. |
Process of Experimentation Court Cases
1. Union Carbide Corp. v. Commissioner, T.C. Memo. 2009-50
The Tax Court upheld Union Carbide’s R&D claims in part because the company provided detailed documentation of its experiments, including test logs, design iterations, and analysis. This case reinforced that a structured, documented process of evaluating alternatives is needed to satisfy the POE requirement.
2. Trinity Industries, Inc. v. United States, 691 F. Supp. 2d 688 (N.D. Tex. 2010)
The court emphasized that POE must involve a methodical plan involving a series of trials to resolve uncertainty. Trinity lost part of its claim because some activities, like minor tooling adjustments, lacked this structure and weren’t backed by credible documentation.
3. Suder v. Commissioner, T.C. Memo. 2014-201
The court allowed R&D credits for projects where the taxpayer demonstrated systematic testing and documentation of efforts to overcome uncertainty in software development. Suder’s success hinged on proving that multiple alternatives were considered and evaluated.
4. Siemer Milling Co. v. Commissioner, T.C. Memo. 2019-37
Siemer Milling’s claim failed because the taxpayer could not show a process of experimentation. The company made process improvements, but they didn’t systematically evaluate alternatives or use trial and error. The court said Siemer’s activities were “routine and did not rise to the level of experimentation.” Moreover, the company did not provide any documentation to support their POE claims.
5. Phoenix Design Group, Inc. v. Commissioner. TC Memo. 2024-113
While the taxpayer satisfied the business component and technological information tests, the court determined that the taxpayer’s engineering activities lacked sufficient uncertainty and investigative efforts and therefore did not meet the process of experimentation requirement.
Best practices for documenting your Process of Experimentation
To satisfy the process of experimentation test, we recommend documenting one or more hypotheses as to how one or more alternatives, techniques, designs, or methodologies might be used to develop a business component, the related tests conducted to evaluate the hypothesis, analysis of the test results, and corrective actions taken, if applicable. It is not sufficient that the taxpayer uses a method of simple trial and error to validate that a process or product change without sufficient documentation.
A formal documentation or design process (e.g. software development life cycle or SDLC) while helpful is insufficient to satisfy the process of experimentation test. Each underlying business component or project must also be documented and reflect the research activities performed.
In a software-based company, this may include architectural documents, design or requirements documents, unit tests, Github commit history, automated or manual test logs, and project management exports. This paints a picture of the activities performed and often ties individuals’ contributions to those activities.
This becomes more challenging for companies using an agile development framework as much of the formal design, development, and evaluation documentation is sacrificed in the name of expediency. We recommend reviewing the recommended documentation with members of the development or engineering teams and evaluating how that level of detail could be captured from the existing process or setting up processes in order to substantiate R&D activities.
By understanding the underlying tax law and judicial interpretation required to satisfy the process of experimentation requirements, taxpayers can have more confidence that their claims will be validated on audit.
Your Next Step
Want to learn more about how Tanner can help with your R&D credit and documentation? Let’s talk.