Six Questions Technology Consulting Firms Ask About the R&D Credit, From Billable Hours to Offshore Teams

TL;DR

• Billing a client for development time does not automatically disqualify it. What often matters is the contract: who bears the risk, and who keeps the rights.
• Reusable accelerators, connectors, and frameworks a firm builds and keeps may be a different story from client-owned deliverables, and billing status alone does not decide whether wages count.
• Offshore work is generally treated differently from U.S. work, and routine platform configuration is generally not research, so sorting your projects early can make a real difference.

Why consulting firms ask these questions

The federal research credit (Section 41 of the Internal Revenue Code) is designed to reward businesses that develop or improve products, processes, software, or techniques through work that involves technical uncertainty and experimentation. It was never limited to labs or product companies. But consulting firms tend to hesitate, because most of their work is done for someone else, on someone else's invoice. Here are the six questions we hear most often.

1. "We bill clients for most of our developers' time. Does that rule it out?"

Not automatically. The regulations exclude research to the extent it is "funded" by another party, and they speak to this directly in Treasury Regulation 1.41-4A. In general terms, two things drive the analysis: whether the payment is contingent on the success of the work, and whether your firm keeps substantial rights in the results under the agreement. If a firm keeps no substantial rights, the work can be treated as fully funded. If it keeps substantial rights, the work may be treated as funded only to the extent of the payments.

Those outcomes depend on the specific language in each agreement, which is why contract review is part of a careful analysis rather than an afterthought. A statement of work that reads one way for a fixed-fee custom build can read very differently for a staffing-style engagement. We would not want to guess at yours from the outside.

2. "Between projects we build reusable accelerators, connectors, and frameworks. Do those count?"

They may be worth a close look. When a firm builds a tool for its own use across many engagements, and keeps the rights to it, the funded-research question can look quite different from client-owned custom work. If the work meets the usual tests, including technical uncertainty and a process of experimentation, it can be a strong candidate.

One distinction to know: software developed mainly for a firm's own general and administrative functions (think internal finance or HR tools) is "internal use software" and faces a higher bar. But under Treasury Regulation 1.41-4(c)(6), software developed to be commercially sold, leased, licensed, or otherwise marketed to third parties, or to let third parties interact with your firm, is not treated as internal use software. Where a given accelerator lands depends on how it is built and used.

3. "What about bench time and non-billable hours?"

The statute looks at what the work was, not whether it was billed. Wages can count when an employee is engaged in qualified research, or in the direct supervision or direct support of it, as described in Section 41. A developer experimenting on an internal accelerator during bench time is not disqualified simply because no client was invoiced. The flip side is also true: billable hours are not automatically eligible. Eligibility turns on the activities performed, supported by records that tie people to projects.

4. "Part of our team is offshore. Is that a problem?"

It can matter. Section 41 excludes research conducted outside the United States, and that rule can apply to work done by contractors as well as employees. In practice, a firm with a blended team usually needs to separate U.S.-based work from offshore work before anything is counted. The U.S. portion may still qualify, depending on the facts. Knowing where the work was performed, and by whom, is an easy record to keep going forward and a harder one to reconstruct later.

5. "Much of our work is implementing platforms. Is that research?"

Often the routine parts are not, and that is worth being clear-eyed about. Standard configuration, data loading, and deployment of an established platform are generally not the kind of work the credit targets. Custom integrations, new architectures, and problems where the right design was not known at the outset, and had to be worked out by testing alternatives, are the areas that may qualify. The regulations do not require a breakthrough: under 1.41-4(a)(3), the work does not need to discover something new to the field, but it does need real technical uncertainty and a process of experimentation. A project mix usually contains both kinds of work, and the value is in telling them apart.

6. "Will a study pull my delivery team off client work?"

This is the objection we hear most, and it is about bandwidth more than eligibility. Strata's initial assessment is $0, and a typical engagement runs roughly four to eight weeks, though timeframes vary and your engagement agreement governs. The time asked of your team is modest, and most of the work sits with us. The finished package is built so your CPA can pick it up and file without starting over. If it would help to see whether any of your project work is worth a look, the first conversation costs nothing.

What to do next

Sort your last year of projects into three piles: client-owned custom builds, reusable tools your firm keeps, and routine implementation. The first conversation with a specialist usually starts there. If you would like a second set of eyes, you can request a no-cost initial assessment. For background on how the IRS reviews these claims, see the agency's Audit Techniques Guide on qualified research activities and the About Form 6765 page.

This post is for informational purposes only and does not constitute tax or legal advice. Consult a qualified tax professional regarding your specific circumstances.