Before a company should be asking “How much credit can we get?”, there’s a more important question to answer:

“Are we actually ready to do an R&D study — and do it well?”

Readiness has very little to do with enthusiasm and a lot to do with whether the right information exists, in the right form, at the right time.

This checklist is designed to help leadership teams pressure-test readiness before committing to a study, setting expectations, or pulling engineers into the process.

1. Is Your R&D Actually Performed in the U.S.?

Start here. This alone disqualifies more claims than people expect.

Ask plainly:

• Where were the engineers physically located while doing the work?

• Which development activities occurred in the U.S.?

• Which occurred offshore?

For federal purposes, only U.S.-performed research qualifies.
Ownership of IP, where the company is incorporated, or where the roadmap was set does not change this rule.

If your development team is split globally, you need a clean separation — not assumptions.

2. Do You Have Complete Payroll Data for 2025?

Payroll is usually the largest driver of the credit.

Before a study starts, you should be able to easily access:

• W-2 wages for U.S.-based employees

• employee names, titles, and roles

• payroll by pay period or month

• bonuses and equity compensation (for context, even if excluded)

If payroll data is fragmented across systems or requires manual reconstruction, that’s a readiness issue — not a tax issue.

3. Can You Clearly Identify Who Was Involved in R&D?

Titles don’t determine qualification — activities do.

You should be able to answer:

• Which engineers were doing technical problem-solving?

• Which non-engineers (QA, DevOps, product, leadership) supported that work?

• Who supervised or directed technical experimentation?

If the answer is “everyone kind of did everything,” that’s normal — but it still needs to be articulated clearly.

4. Do You Have a Record of What R&D Actually Happened in 2025?

This doesn’t mean perfect documentation.

It means there is a defensible trail showing:

• what was being built or improved,

• where technical uncertainty existed,

• what didn’t work the first time,

• and how problems were ultimately resolved.

In practice, this usually lives in:

• Jira or other ticketing systems

• Git commits and pull requests

• design docs or architecture notes

• test failures, bugs, and refactors

If none of that exists — or it exists but can’t be accessed — the study will be harder and riskier.

5. Are Your R&D Vendors and Contractors Clearly Tracked?

Contractor costs are commonly misunderstood.

Before starting a study, you should know:

• which vendors performed technical work,

• whether they were U.S.-based,

• whether the company retained rights to the work,

• and how those costs are recorded in the general ledger.

If vendor spend is mixed between R&D and non-R&D services with no clarity, that needs to be addressed upfront.

6. Do You Have Prior-Year Tax Returns Available?

This is non-negotiable for a proper estimate.

Prior-year returns are needed to:

• understand historical R&D patterns,

• evaluate the base period or “hurdle,”

• assess carryforwards,

• and set realistic expectations.

Without prior-year returns, any estimate is guesswork — and often misleading.

7. Do You Understand Whether 2025 Was a Growth Year or a Down Year?

This matters more than most people realize.

Ask:

• Did R&D investment increase meaningfully in 2025?

• Or was it flat or down relative to prior years?

The R&D credit rewards increases over time, not absolute spend.
Understanding this upfront prevents disappointment later.

8. Are You Prepared for Timing — Not Just the Credit Amount?

Before starting a study, leadership should align on:

• whether credits will be used immediately or carried forward,

• whether payroll offsets apply,

• and how long realization may take.

If the expectation is fast cash, that needs to be reset early.

9. Do You Have the Right People Available for Light Involvement?

A good study should not be disruptive — but it’s not zero-touch.

You’ll likely need:

• limited time from engineering leadership,

• someone who understands how work flowed in 2025,

• and a finance contact who can provide data promptly.

If everyone is unavailable or disengaged, results suffer.

10. Are You Treating This as a One-Off or a System?

Finally, step back and ask:

• Is this a one-time exercise?

• Or something we may want to repeat year over year?

That answer influences:

• methodology,

• documentation rigor,

• and how much long-term value the credit actually delivers.

Companies that view the credit as a system tend to extract more value — with less stress — over time.

The Takeaway

Being “eligible” for the R&D credit is not the same as being ready for an R&D study.

Readiness is about:

• data availability,

• clarity of work performed,

• realistic expectations,

• and internal alignment.

If most of the boxes above are checked, a study is likely to go smoothly and produce defensible results.

If several are missing, the best move is usually to pause, prepare, and then proceed — not rush.