Every audit starts with an information document request, and lately those requests look different than they used to. Exam teams no longer accept a summary spreadsheet of qualified research expenses as sufficient support. They want the underlying project record: who worked on it, what technical uncertainty they faced, and how the wage allocation ties back to that specific work.
We detailed the current pattern in our review of what IRS exam teams are actually requiring in R&D credit audits right now. This post goes a step further and looks at how a Fortune 100 tax team builds a defensible position before the request ever arrives.
The Shift From Summary to Substantiation
For years, many organizations supported R&D credit claims with high-level narratives and aggregate QRE totals. Examiners accepted that approach because they had no easy way to challenge it component by component. That has changed. Between Form 6765’s new Section G requirements and heightened scrutiny on refund claims, examiners now expect the same level of detail the form itself demands.
This means your defense file and your filing need to match from the start. If Section G requires project-level detail, your workpapers need that same detail sitting behind them, not a reconstruction assembled after the audit letter shows up.
What Examiners Consistently Request
Technical narratives written by the people who did the work top the list, not summaries written secondhand by finance. A narrative written by an engineer describing the specific uncertainty they faced carries far more weight than a paraphrased version drafted months later by someone who wasn’t on the project.
Contemporaneous time tracking matters just as much. Estimates built after the fact, however reasonable they seem, draw more scrutiny than records created while the work was happening. If your organization still relies on year-end interviews to reconstruct time allocation, that process alone adds to your audit risk before an examiner has even looked at your numbers.
Examiners also want a clear line from technological uncertainty to the four-part test. They increasingly push back on activities that read as routine engineering rather than actual experimentation. The bar for demonstrating a process of experimentation has risen over the past few years, and generic language no longer satisfies it the way it might have five years ago.
Where Funded Research Questions Enter the Picture
Funded research disputes have become a recurring audit theme, particularly for organizations that perform contract research or receive third-party funding for development work. Our analysis of Smith v. Commissioner and its clarification of funded research rules shows how the retained-rights question can determine whether an entire research program qualifies. If your organization has funding arrangements tied to research activity, revisit those agreements now, well before an examiner asks you to explain them under pressure.
Building Your Pre-Audit Position
Don’t wait for the information document request to discover your gaps. Run an internal mock audit against your own strongest and weakest business components. Pull the documentation you’d hand an examiner today, and take an honest look at whether it would survive the same scrutiny an examiner would apply to it.
Focus first on your largest-dollar business components, since those draw the most examiner attention. If your top five components by QRE can’t produce clean technical narratives and time records, that’s where your risk concentrates, and that’s where remediation should start rather than spreading effort evenly across every component.
It’s also worth considering whether your existing claim would benefit from an independent review before you file, rather than after an audit letter arrives. Our reverse audit work is built around this exact scenario: finding weak spots in a claim while you still have time to fix them, instead of discovering them for the first time in an examiner’s request.
When the Audit Letter Does Arrive
If you’re already facing an examination, the priority shifts from prevention to defense. Organizations that bring in tax controversy support early in the process, rather than after the first unfavorable exam finding, tend to preserve more of the credit’s value. Early engagement gives you room to correct course, supplement weak documentation, and shape the technical narrative before positions on either side become harder to move.
Where This Leaves Tax Teams Going Into 2026 Filing Season
Examiners have raised their standard for what counts as adequate substantiation, and that standard isn’t going back down. Fortune 100 tax teams that match it ahead of time, rather than scrambling to meet it after an audit letter arrives, will spend far less time and money defending claims that were solid from the start. The documentation is what proves the underlying research was real, thorough enough that an examiner doesn’t have to take your word for it.
Where to Start This Quarter
Pick one business component you’d consider high risk and one you’d consider low risk, then pull the full documentation file for each. Compare both against what an examiner would ask for under the current standard. The gaps you find in that exercise, applied across your full portfolio of components, will tell you roughly how much work this transition still requires. If you’re already facing an exam letter, or want a second opinion before you get one, get in touch and we’ll help you think it through.