Research and development is the planned, creative work a business does to build new knowledge and turn it into a new or improved product, service or process. To count as research and experimental development in the federal statistical definition, an activity must satisfy five core criteria: it has to be novel, creative, uncertain, systematic, and transferable or reproducible (NSF NCSES, Definitions of Research and Development). Three separate US rule sets then run over the money spent on it: one governing the financial statements, one governing the deduction on the tax return, and a third governing the research credit.
The Research Half and the Development Half
The statistical taxonomy splits the activity three ways, into basic research, applied research and experimental development (NSF NCSES, Definitions of Research and Development). The accounting standard splits it two ways and defines each side separately. Research is planned search or critical investigation aimed at discovery of new knowledge, with the hope that the knowledge will be useful in developing a new product or process or in bringing about a significant improvement to an existing one. Development is the translation of research findings or other knowledge into a plan or design for a new or significantly improved product or process, and it takes in conceptual formulation, the design and testing of product alternatives, the construction of prototypes and the operation of pilot plants (FASB Statement No. 2, paragraph 8).
The same definition says what sits outside the term. Routine or periodic alterations to existing products, production lines and manufacturing processes are not development, even where those alterations are real improvements, and neither market research nor market testing counts.
The Book Rule Is Expense as Incurred
Research and development costs are charged to expense when incurred, and the total charged to expense has to be disclosed for each period an income statement is presented (FASB Statement No. 2, paragraphs 12 and 13).
The refusal to capitalize was reasoned rather than assumed, and the Board’s own basis for conclusions leads with two findings. There is normally a high degree of uncertainty about the future benefits of individual projects, although the element of uncertainty may diminish as a project progresses, and a direct relationship between research and development costs and specific future revenue has generally not been demonstrated, even with the benefit of hindsight (FASB Statement No. 2, paragraphs 39 and 41).
One line inside the standard catches preparers out. Equipment and facilities acquired or constructed for research and development that have alternative future uses are capitalized as tangible assets, and it is the depreciation on them that becomes a research and development cost; where there is no alternative future use, the whole cost is a research and development cost when incurred (FASB Statement No. 2, paragraph 11).
The Tax Side Uses a Different Term
On the return the equivalent term is a research or experimental expenditure, and the population is not the book number relabeled. The regulation defines the tax term as expenditures incurred in connection with the taxpayer’s trade or business which represent research and development costs in the experimental or laboratory sense, which it ties to activities intended to discover information that would eliminate uncertainty concerning the development or improvement of a product. It then excludes a named list, among them quality control testing, efficiency surveys, management studies, consumer surveys, and advertising or promotions (26 CFR 1.174-2, paragraphs (a)(1) and (a)(6)).
The two populations diverge on real items. The cost of obtaining a patent, including attorneys’ fees expended in making and perfecting a patent application, is a research or experimental expenditure (26 CFR 1.174-2(a)(1)), while legal work in connection with patent applications sits on the accounting standard’s list of activities typically excluded from research and development (FASB Statement No. 2, paragraph 10).
The timing splits on where the work is performed. Section 174A(a) allows a deduction for domestic research or experimental expenditures paid or incurred during the taxable year, while section 174 now reaches only expenditures attributable to foreign research, which stay on amortization, for amounts paid or incurred in taxable years beginning after December 31, 2024 (IRS, Rev. Proc. 2025-28). The elections that sit on top of that, the transition rules, and the subsection that sweeps software development into the term are worked through in outsourced accounting for SaaS.
Three Rule Sets Over the Same Work
Book research and development, the deduction for research or experimental expenditures, and the research credit are three separate tests run over one body of work, and qualified research for credit purposes has to satisfy its own statutory definition. Its first prong now runs off the domestic deduction, research with respect to which expenditures are treated as domestic research or experimental expenditures under section 174A, and three further requirements sit on top of it, together known as the four-part test (IRS, Instructions for Form 6765).
So a cost can be research and development on the financial statements, a research or experimental expenditure on the return, and still fail the credit’s qualification test. What staffing a credit claim actually takes is set out in R&D tax credit services, and how a credit differs from a deduction is in tax credit.
Accountably places trained offshore accountants and tax preparers inside US CPA and EA firms, and a firm can run a real block of its own work through a trial task before anything client facing depends on it.
