Documenting a process of experimentation is central to substantiating the process-of-experimentation element of qualified research under Section 41. The Treasury Regulations describe a process of experimentation as an evaluative process designed to evaluate one or more alternatives to resolve a technical uncertainty, involving identifying the uncertainty, identifying alternatives, and conducting an evaluative process. This page explains how to document that process. It is educational and is not individualized advice. For the underlying framework, see our page on process of experimentation.
The Three Components of a Process of Experimentation
The Treasury Regulations describe a process of experimentation as involving three components:
- Identifying the uncertainty — the technical question about capability, method, or design.
- Identifying alternatives — one or more approaches intended to eliminate the uncertainty.
- Conducting an evaluative process — testing, modeling, simulating, or otherwise evaluating the alternatives.
Documenting a process of experimentation means recording each of these components. For more on the first two, see our pages on documenting technical uncertainty and documenting alternatives.
Documenting the Evaluative Process
The third component — the evaluative process — is what distinguishes a process of experimentation from a single predetermined path. Useful documentation of the evaluative process may include:
- Hypotheses or alternatives — what was being evaluated.
- Trials — what tests were conducted, under what conditions.
- Modeling and simulation — inputs, outputs, and results of any modeling or simulation.
- Test results — the data collected from each trial or test.
- Iteration — how results informed the next step or the next alternative.
- Failures — tests that did not produce the desired result and what was learned.
Modeling and Simulation
The regulations expressly identify modeling and simulation as examples of evaluative processes. Where a taxpayer models or simulates alternatives, records of the model inputs, assumptions, outputs, and how the results informed the evaluation can help support the process. Modeling and simulation are not required, but where they are used, documenting them strengthens the record.
Systematic Trial and Error
The regulations also identify "systematic trial and error methodology" as an example. The key word is "systematic." Records should show that the trial and error was structured — identifying the uncertainty, identifying alternatives, and evaluating them in a systematic way — rather than random or ad hoc. A record that shows the sequence of trials, the rationale for each, and how results informed the next step can help demonstrate a systematic process.
Documenting Iteration and Failures
Iteration and failures can be consistent with a process of experimentation. The regulations do not require success, and experimentation often involves tests that do not produce the desired result. Records that show how a failed test of one alternative informed the evaluation of another can help demonstrate a genuine evaluative process. Discarding or not recording failures can leave a gap in the story of how the uncertainty was addressed. For more, see our page on documenting failed prototypes.
Hypothetical Example
Consider a manufacturer that is developing a new heat-treatment process for a new alloy. The records show: the uncertainty (what temperature/time profile can achieve the required hardness), the alternatives (three candidate profiles), the evaluative process (test specimens prepared for each profile, hardness tested after treatment), the results (hardness achieved for each profile), the iteration (one profile modified based on results and retested), and the failures (one profile did not achieve the required hardness and was rejected). This documentation — showing the full evaluative process — can help support the process-of-experimentation element.
This example is illustrative only.
Documentation That May Help
Records that can help support the process-of-experimentation element include experiment logs, test plans, test results, modeling and simulation records, iteration records, and records of failures. Records created during the evaluation tend to be most useful. For more, see our page on R&D tax credit documentation.
Key Takeaway
Documenting a process of experimentation involves recording the hypotheses or alternatives, the trials, modeling or simulations, test results, iteration, and failures. This supports the process-of-experimentation element of qualified research. Contemporaneous records showing the full evaluative process tend to be most useful. Because documentation adequacy is fact-specific, professional review is appropriate.