Ricerca

R&D Tax Credit for Architects and Architecture Firms

Some of your design hours are research. Most of them are not.

Architecture firms can claim the federal R&D credit for design work that resolves a genuine technical uncertainty through experimentation, such as performance modelling, new assemblies, and novel structural or material systems. Style and aesthetics are excluded by statute, and your client agreements decide whether the rest can be claimed. Each project is tested against the four-part test.

Photo by Marsumilae on Unsplash

Can architecture firms claim the R&D tax credit?

Yes, for the part of the work that is technical. The IRC §41 credit reaches work on a business component where the information available did not establish whether the design could be achieved, how to achieve it, or what the right design was, and where the firm resolved that by evaluating alternatives through modelling, simulation, or systematic trial and error.

That is why a facade assembly outside the tested range, a mass-timber floor for a span you had not done before, or a design iterated against a calibrated energy model are all candidates, and why a concept scheme, a permit set, or a code check are not. Architecture carries one exclusion other design disciplines barely notice: §41(d)(3)(B) says research is never for a permitted purpose if it relates to style, taste, cosmetic, or seasonal design factors.

The second half of the problem is contractual. Almost every hour is billed to a client, so the funded-research exclusion has to be answered agreement by agreement. Fee structure is a starting position, not an answer, and the reported outcomes for design firms cut both ways. Engineering firms face the same contract question with different technical work; that is covered on our R&D tax credit for engineering firms page.

Illustrative examples

Seven situations that often hide qualified research

Illustrative situations, not client work. If one reads like a project you ran, there may be a qualified business component underneath it.

  • Example 1

    A facade that would not hit the energy target as drawn

    Glazing ratios, shading depths, and wall build-ups were modelled and re-modelled against a calibrated energy model until the predicted performance held, and the elevation changed because the numbers said so.

    Why it can qualify: Improving performance is a permitted purpose, and iterating alternatives against a calibrated simulation is a process of experimentation.

  • Example 2

    A mass-timber floor at a span the studio had never detailed

    Panel depths, composite toppings, and support conditions were analysed with the engineer against a vibration criterion, and two schemes were dropped on the numbers before one was carried.

    Why it can qualify: Uncertainty about the appropriate design of a new structural assembly, resolved by evaluating alternatives rather than by selecting a known detail.

  • Example 3

    An embodied-carbon target that ruled out the usual structure

    Low-carbon concrete mixes and a hybrid framing option were evaluated with the engineer and the supplier, and a test panel was poured before the specification was fixed.

    Why it can qualify: Capability and method uncertainty in the physical sciences. A sustainability target is a performance requirement when meeting it takes experimentation.

  • Example 4

    A shading study that kept failing the glare criterion

    A parametric model generated and simulated dozens of fin geometries against measurable daylight and glare metrics, and the design moved on the results, not on preference.

    Why it can qualify: Computational exploration against measurable performance criteria is modelling and simulation in the regulatory sense. Choosing among options that all work on looks is not.

  • Example 5

    A lightweight wall that missed the acoustic separation

    Three assemblies were developed, a mock-up was built, and field measurements sent the design back once before the rating held.

    Why it can qualify: A performance requirement the available information could not meet, resolved by mock-up and measurement. The mock-up materials consumed are supplies.

  • Example 6

    A room module that had to survive transport and craning

    A prefabricated room module needed new connection and tolerance details, so prototypes were built, lifted, and inspected before the design went to the fabricator.

    Why it can qualify: Design uncertainty resolved by prototyping. The first module is the candidate; repeating the proven module on the next site is duplication.

  • Example 7

    An adaptive-reuse scheme resting on a floor nobody could rate

    The existing slab’s capacity had to be established by investigation and analysis before a new use could be planned on it, and the first strengthening concept did not survive the second round of numbers.

    Why it can qualify: The information available did not establish the method. Establishing it by test and analysis is what the statute asks for.

Illustrative situations, not client work. Whether any of them qualifies for you depends on your facts, your agreements, and your evidence.

A pencil and ruler resting on a dimensioned technical drawing
A dimensioned drawing under revision. Illustrative.Photo by Sven Mieke on Unsplash

Documentation that works for design firms

A studio already produces most of the evidence a claim needs, in a form nobody files for tax: option studies with the schemes that were rejected, model runs and their assumptions, superseded detail revisions, mock-up photographs and test results, and the design-meeting minute where the answer changed.

What is almost never in the record is the allocation: which hours went to the uncertain part of the project, and which went to documenting a settled design. Project accounting by phase and task usually gets you most of the way, because it already separates schematic design from documents and construction administration. A flat percentage applied across every project cannot answer the question, and it is the first thing an examiner asks about.

How substantiation is assembled

Architectural design work that can qualify

Representative activities that can meet the four-part test when the outcome was genuinely uncertain and alternatives were evaluated.

Energy and performance modelling

Iterative design against a calibrated simulation, where alternatives are tested and the design changes on the result.

High-performance envelope and facade assemblies

Thermal, moisture, air, and durability performance of assemblies that fall outside tested and published ranges.

New materials and structural systems

Mass timber, low-carbon concrete, and hybrid systems where the appropriate design has to be established by analysis and test.

Computational and parametric design

Generative exploration scored against measurable criteria such as daylight, glare, structure, or thermal comfort.

BIM-driven design experimentation

Using the model to test alternatives against measurable criteria before the design is fixed, as distinct from producing documents from it.

Sustainability and net-zero targets

Embodied-carbon, energy, and water targets that cannot be met with a known approach and have to be reached by experimentation.

Acoustic and vibration performance

Separation, isolation, and floor-vibration criteria that standard assemblies do not reach, resolved by modelling, mock-up, and measurement.

Prefabrication and modular systems

Connections, tolerances, and transport behaviour of prefabricated components developed by prototype and test.

Design-phase mock-ups and testing

Physical and instrumented mock-ups built during design to answer a question the drawings could not answer.

What qualifies, and what does not

Architectural work that can qualify for the R&D credit compared with work that does not
Can qualify Does not qualify
Developing a facade assembly outside the tested range and proving it by analysis and mock-upChoosing a cladding material, colour, or pattern for how it looks
Iterating the design against a calibrated energy model until a performance target holdsRunning a compliance model to document that a finished design meets the code baseline
Evaluating structural or material alternatives for a new assembly, with the engineerMassing studies, renderings, and aesthetic iteration
Parametric exploration scored against measurable daylight, glare, or thermal criteriaProducing construction documents, permit sets, and code checks for a settled design
Prototyping the first prefabricated module or connectionRepeating a proven prototype building on a new site, or adapting it to one client’s brief
Resolving a field condition that forces a genuine redesignConstruction administration: submittals, RFIs, and site observation
General and illustrative. Qualification is decided per business component on your facts, and qualifying work can still be excluded as funded research under §41(d)(4)(H).

Typical QRE categories for an architecture practice

What spending counts toward the credit, for a business whose costs are almost entirely people.

Typical QRE categories and their statutory basis
Expense category What goes into the base
Technical wages§41(b)(2)(A)-(B)W-2 wages for architects, designers, computational designers, and modellers for time spent on qualified design work, plus direct supervision and direct support of it.
Contract research (65%)§41(b)(3)65% of amounts paid to U.S. consultants and testing laboratories (energy modellers, facade and acoustic consultants, test labs) for qualified research performed on your behalf, under an agreement entered into before the work that requires you to pay even if the research is not successful.
Supplies§41(b)(2)(C)Materials consumed in mock-ups, test panels, and physical prototypes built during design. Never depreciable equipment.
Computer rental§41(b)(2)(A)(iii)Amounts paid for the right to use computers in qualified research, such as rented compute for a large simulation run. Per-seat software licences are not a §41 expense category.
General and illustrative. Only qualified research performed in the United States, Puerto Rico, or a U.S. possession is eligible, and contract research enters the base at 65% of the amount paid under §41(b)(3).

What the base usually looks like

Illustrative

A directional shape for an architecture practice, not a benchmark. Wages carry nearly the whole claim, which is why the hour-level allocation has to be defensible rather than assumed.

Technical wages - Architect, designer, and modeller time on projects with a real technical unknown.
84%
U.S. contract research - Performance consultants and test labs you pay whatever the outcome, in the base at 65%.
12%
Supplies - Mock-up, test-panel, and prototype materials consumed during design.
3%
Computer rental - Rarely material, and software seats do not count.
1%

Where the line sits

Consultant fees count at 65% only when you pay them whatever the outcome and the research is performed on your behalf. A consultant billed through to the client at cost is usually the client’s expense, not yours.

Seat licences for BIM, modelling, and rendering software are not a §41 expense category, however large the invoice. Neither is rent or the overhead loaded into your billing rate: a fully loaded cost model will overstate the credit every time.

Full QRE rules, category by category

Exclusions to watch

Where architecture-firm claims fail

One of these is about the contract and the rest are about the hours. A claim that ignores either half is the kind that does not survive an examination.

§41(d)(3)(B)

Style, taste, and appearance

Research that relates to style, taste, cosmetic, or seasonal design factors is outside the permitted purposes by statute. Massing, composition, material palettes, and aesthetic iteration are the practice of architecture, not qualified research. Where an appearance decision created a real structural, envelope, or acoustic problem, it is the technical work that resolved it that becomes the candidate.

§41(d)(4)(H)

Work the client funded

Research is funded, and excluded, unless you both bear the financial risk of failure and retain substantial rights in the results. Payment that does not depend on success is funding, and so is an agreement that leaves you no substantial rights. Nearly every hour in a practice is billed to a client, so this is answered agreement by agreement before anything enters a base.

§1.41-4(a)(3), (a)(5)

Documents, permitting, and code compliance

Construction documents for a settled design, permit sets, and routine code compliance resolve no technical uncertainty and involve no process of experimentation. On most projects this is the majority of the fee, which is why an honest claim is a fraction of the hours rather than a percentage of the practice.

§41(d)(4)(A), (B), (C)

Construction, adaptation, and repeat prototypes

Work after the design is ready for use (construction administration), adapting an existing design to a particular client’s requirement, and rolling a proven prototype building onto a new site are each excluded.

§41(d)(4)(D), (G)

Programming, surveys, and design research in the arts

Occupant surveys, post-occupancy studies, space programming, and market work are surveys and studies, and research in the social sciences, arts, or humanities is excluded outright. Both can sit next to qualifying work on the same project.

§41(d)(4)(F)

Offshore production support

Research performed outside the United States, Puerto Rico, or a U.S. possession is excluded regardless of who employs the staff. Practices using offshore modelling or documentation support need that time screened out, and the related costs are foreign research for §174 purposes too.

The contract question

Fixed fee, hourly, and who owns the drawings

Under §41(d)(4)(H) and Treas. Reg. §1.41-4A(d), research is funded unless you bear the financial risk of failure and keep substantial rights in the results. Both prongs are read from your agreements.

How common architecture fee structures start on the funded-research question
Fee structure Starting position What actually decides it
Stipulated sum or fixed feeStrongest starting position. Extra design effort is not paid for, so you carry the cost if the work does not succeed.Whether payment is tied to meeting objective requirements (acceptance, rejection, rework at your cost), not just a standard of care. A fixed fee alone was not enough in 2024.
Percentage of construction costCloser to a fixed fee than to hourly, because the fee does not move with your design hours.The same payment, acceptance, and rework terms as a fixed fee. Read the agreement, not the fee basis.
Hourly or time and materialsWeakest starting position. The client pays for effort whatever the outcome.If you keep substantial rights, the research is funded to the extent of the payments you become entitled to, so little is usually left.
Hourly with a not-to-exceed capWeak. A capped cost-reimbursement agreement was held funded in Geosyntec (11th Cir. 2015).Whether any payment is conditioned on success at all. A cap limits the client’s cost; it does not make payment depend on results.
General information, not legal advice. The analysis runs agreement by agreement, reading every agreement between the parties together.

Risk: what the courts have looked at

In Meyer, Borgman & Johnson v. Commissioner (8th Cir. 2024), a structural engineering firm lost on fixed-price contracts. Performing with professional skill and care and complying with codes was proper performance, not successful performance, and the agreements had no express terms making payment depend on success, such as rejection of non-conforming work or a refund if benchmarks were missed.

In Populous Holdings v. Commissioner (T.C. Dkt. No. 405-17), the Tax Court granted summary judgment in December 2019 to an architectural design firm on fixed-price contracts: clients paid fixed prices for final products, so payment depended on the firm’s successful performance. The order did not decide whether the design work was qualified research, and as an unpublished order it is persuasive reasoning only.

Rights: who can use what you figured out

Substantial rights do not mean exclusive rights. In Populous, the clients’ ownership of the delivered documents was not fatal, because nothing in the contracts stopped the firm from using the research it had performed or required it to pay for that use.

The standard AIA owner-architect agreement (B101-2017) is a useful reference point: the architect is deemed the author and owner of its Instruments of Service and keeps copyright, and the owner receives a license limited to the project. Its standard-of-care language is the kind of obligation the Eighth Circuit said does not by itself put payment at risk. Many practices sign modified or owner-drafted forms, so the analysis follows the signed agreement, not the template.

Case outcomes for design firms have gone both ways on facts that look similar from the outside, which is the honest reason to read the agreements rather than rely on a rule of thumb. Our note on funded research and contract clauses works through the cases in more detail.

The federal credit is rarely the whole number

Most states with a corporate income or franchise tax run their own R&D credit off a similar research base, each with its own rate, cap, carryforward, and refundability. For a practice with studios in several states, where the qualified work was performed matters to both numbers.

Two deductions sit next to the credit. §174A restored immediate expensing of domestic research and experimental costs for tax years beginning after December 31, 2024, on a definition broader than the four-part test. And the §179D deduction that government and tax-exempt building owners could allocate to designers does not apply to property whose construction begins after June 30, 2026, so only projects that started construction by then are still in play.

What we ask for first

  • The owner agreements on your largest projects, in full, with amendments
  • Project accounting by phase and task, not just by job
  • Option studies, model runs, and superseded detail revisions
  • Consultant agreements, and whether their fees are billed through
  • Which work was supported offshore, and how much

Summary only - the analysis is done per agreement and per business component, not per firm.

What an architecture-firm study can look like

A hypothetical scenario to show how the pieces fit together. It is not a quote, projection, or promise of results.

~35-person architecture practice
Illustrative
Technical payroll
$3.2M
Projects surviving the funded-research screen
~50%
Share of those hours on genuinely uncertain work
~15%
U.S. performance consultants you bore the cost of
$60K
Estimated QRE
~$280K
Illustrative federal credit
≈ $17K-$28K

Plus the §174A deduction on domestic research and experimental costs, on a definition broader than the credit’s.

Illustrative only. Figures are hypothetical and rounded; consultant fees are included at 65%. The federal credit commonly works out to roughly 6-10% of QRE depending on method, filing history, and the §280C election. Note how much the two screens remove: that is what an honest architecture-firm claim looks like. This is not a quote or a guarantee.

Don’t forget §174A

The deduction reaches further than the credit

IRC §174A restores immediate, full expensing of domestic research and experimental costs for tax years beginning after December 31, 2024, on a definition broader than the §41 four-part test. For a design practice that gap matters: costs that do not clear the credit can still be §174A expenditures.

R&D tax credit for architects - frequently asked questions

Can architects claim the R&D tax credit?
Yes, for some of their work. The IRC §41 credit reaches design that is technological in nature, relying on the principles of engineering or the physical sciences, and that resolves a genuine uncertainty about capability, method, or the appropriate design through a process of experimentation. Performance-driven design, new assemblies, and novel structural or material systems can meet that. Aesthetic design cannot, and client agreements decide whether the qualifying hours can be claimed at all. The practical question on any project is which hours were resolving a technical unknown; the honest answer is usually a minority of them.
Does conceptual or aesthetic design count?
No. §41(d)(3)(B) says research is never treated as conducted for a permitted purpose if it relates to style, taste, cosmetic, or seasonal design factors. Concept design, massing, and material selection for appearance are outside the credit however difficult they were. If an appearance decision created a technical problem, such as a cantilever, a glazing ratio, or an acoustic condition that standard details could not handle, the work that resolved that problem is analysed on its own.
Our agreements are stipulated sum. Does that settle the funding question?
No. A fixed fee is the best starting position, but in 2024 the Eighth Circuit affirmed the Tax Court against a structural engineering firm on fixed-price contracts (Meyer, Borgman & Johnson). Obligations to perform with professional skill and care and to comply with codes described proper performance, not successful performance, and the agreements had no express terms making payment depend on success. An earlier, unpublished Tax Court order in Populous Holdings (December 2019) went the other way for an architectural design firm on fixed-price contracts. The difference is in the words of each agreement - see how contract clauses decide who gets the credit.
We bill most projects hourly. Are we out?
Mostly, for those projects. When a client pays for your hours whatever the outcome, you are not bearing the risk of failure, and capped cost-reimbursement agreements were held funded in Geosyntec (11th Cir. 2015). Under Treas. Reg. §1.41-4A(d)(3), where you retain substantial rights the research is funded only to the extent of the payments you become entitled to by performing it, so the question narrows to the amounts. Firms with a mix of fee types usually find the claim concentrates in their fixed-fee work.
The client owns the drawings. Have we given away substantial rights?
Not necessarily. Substantial rights do not mean exclusive rights: the Federal Circuit has held that a performer keeps substantial rights where it may use the research results in its business without paying for that use. The standard AIA owner-architect agreement (B101-2017) treats the architect as the author and owner of its Instruments of Service, with copyright, and gives the owner a license limited to the project. In Populous, client ownership of the delivered documents was not fatal because nothing stopped the firm from using the underlying research. That order is persuasive reasoning only; a sweeping assignment of all work product and know-how with nothing carved back is the risk.
Is energy modelling qualified research?
It can be, when the model is being used to resolve a genuine uncertainty rather than to document a decision already made. Calibrating a model, testing alternatives against it, and changing the design because the results said so is experimentation. Running a compliance model to demonstrate that a finished design meets a code baseline, or assembling a certification submission, is documentation and is generally outside the credit.
Does BIM or computational design count?
The software does not; what you do with it might. Producing and coordinating documents in a model is not experimentation. Using the model or a parametric script to generate and evaluate alternatives against measurable performance criteria, where the outcome was not known in advance, can be. BIM and analysis software seat licences are not a §41 expense category either way.
What documentation do architecture firms need?
The regulations do not prescribe a format, but the claim has to be substantiated by business component. For a studio that usually means option studies with the rejected schemes, model runs and their assumptions, mock-up photographs and test results, design-meeting minutes where the answer changed, and project accounting by phase and task that ties hours to those components. A flat percentage applied across every project is the first thing an examiner questions. More in documentation and substantiation.
Does §174A apply to an architecture practice?
It can, and its reach is wider than the credit’s. §174A restores immediate expensing of domestic research and experimental expenditures for tax years beginning after December 31, 2024, on a definition broader than the §41 four-part test, while foreign research stays on 15-year amortization. Costs that fall outside the credit can still be §174A expenditures, so the two populations are identified separately - see our Section 174A guide and how §174A and §41 differ.
What about the §179D deduction for designers?
§179D is a separate deduction for energy-efficient commercial building property, and a government or tax-exempt building owner can allocate it to the designer. The 2025 reconciliation law (P.L. 119-21) ended it for property whose construction begins after June 30, 2026. Projects that began construction on or before that date may still carry an allocation, so check the construction timing before relying on one. It does not change the §41 analysis.

Next

Funded research: how your contracts decide who gets the credit

The risk and rights prongs worked through with the cases, including the design-firm outcomes on both sides.

The test itself

The four-part test, applied the way an examiner applies it

Permitted purpose, technological in nature, uncertainty, and experimentation, with pass and fail examples.

For engineering firms

R&D tax credit for engineering firms

Structural, civil, MEP, and product engineering, with the same contract question on the other side of your consultant agreements.

New to the credit? Start with the R&D tax credit overview, or put rough numbers into the R&D credit calculator.

Find the design hours that would actually survive a review

Tell us what your practice designs and how the work is contracted, and we’ll separate the genuinely uncertain hours from aesthetics and documentation - screening every agreement for funded research, and capturing §174A - reviewed and finalized by R&D experts and backed by Audit Protection. Contact us for pricing tailored to your study.

[email protected] We typically reply within one business day.
Get your free R&D credit assessment

We typically reply within one business day.