Documenting failed prototypes and experiments is an important part of substantiating an R&D tax credit claim. The Treasury Regulations provide that a taxpayer need not succeed in developing or improving a business component, and failed tests can be evidence of a process of experimentation. This page explains how to document failures in a way that can help support a credit claim. It is educational and is not individualized advice. For the underlying framework, see our page on failed experiments.
Failures Can Be Evidence of Experimentation
A common misunderstanding is that failures disqualify a credit claim. They do not. The regulations provide that a taxpayer need not succeed in developing or improving the business component. Failed tests and unsuccessful iterations can be consistent with a process of experimentation, and records of them can be important context. A failed test of one alternative that informed the evaluation of another can help demonstrate a genuine evaluative process. For more, see our page on failed experiments.
What to Record for Each Failure
For each failed prototype or experiment, useful documentation may include:
- What was tested — the specific design, material, method, or configuration that was tested.
- Why it was expected to work — what made it a candidate for resolving the uncertainty.
- What the failure was — what happened (did not achieve the target, broke, did not perform as expected).
- Why it failed — the root cause or explanation, if known.
- What changed — what was modified for the next iteration based on the failure.
- What the next iteration was — the next alternative or approach tested.
Records that capture all six elements tend to be more useful than records that simply note that a test failed.
Why Failures Are Often Forgotten
Failures are often forgotten after the fact, particularly when a project eventually succeeds. The successful final design is remembered; the failed alternatives that led to it may not be. This is why contemporaneous records of failures are valuable — they capture the evaluative process as it actually occurred, including the alternatives that did not work. A record that shows the team tested and rejected several alternatives before finding one that worked can help demonstrate a genuine process of experimentation.
Hypothetical Example
Consider a manufacturer that is developing a new welding process for a new alloy. The first weld parameter combination tested produced cracks; the records show what was tested, why it was expected to work, what the failure was (cracking), why it failed (heat input too high), what changed (heat input reduced), and what the next iteration was (a modified parameter set). The second iteration produced insufficient penetration; the records show the same six elements. The third iteration achieved the required result. This documentation — showing the full sequence including failures — can help support the process-of-experimentation element.
By contrast, a record that says only "we developed a welding process" without showing the failed iterations is less useful.
This example is illustrative only.
Documentation That May Help
Records that can help support failure documentation include experiment logs, test results, failure analysis records, design iteration records, and project communications discussing failures and next steps. Records created during the evaluation, when the failures occurred, tend to be most useful. For more, see our page on R&D tax credit documentation.
Key Takeaway
Documenting failed prototypes and experiments involves recording what was tested, why it failed, what changed, and what the next iteration was. Failures can be evidence of a process of experimentation and do not disqualify a claim. Contemporaneous records of failures tend to be most useful. Because documentation adequacy is fact-specific, professional review is appropriate.