Last verified: August 2026. This is general information about how the SR&ED program works, written for research administrators, university-industry liaison offices, and technology transfer staff who encounter SR&ED alongside grant funding — it is not tax advice, and no claim should be filed without a qualified tax professional and, where applicable, the corporation’s own SR&ED consultant reviewing the specific facts.
TL;DR: The Scientific Research and Experimental Development (SR&ED) program is Canada’s largest single source of federal support for business-performed R&D, delivered as a tax incentive rather than a grant — administered jointly by the Canada Revenue Agency (CRA), which assesses eligibility and expenditures, and the Department of Finance, which sets policy. It works through two mechanisms: SR&ED expenditures are deductible against income, and a portion also earns an investment tax credit (ITC). Canadian-controlled private corporations (CCPCs) below a taxable-capital threshold can get up to 35% of qualifying expenditures back as a refundable credit — cash even with no tax payable — on an enhanced-rate expenditure limit that Bill C-15 raised from $3 million to $6 million (Royal Assent March 26, 2026, effective for tax years beginning after December 15, 2024). Other corporations, and CCPC spending above the limit, earn a 15% non-refundable credit. Eligibility does not turn on the field of research or the outcome — it turns on whether the work meets three tests CRA applies project by project: scientific or technological uncertainty, systematic investigation, and technological advancement. For institutions with industry-sponsored research, spin-out companies, or collaborative research agreements, SR&ED interacts directly with other government funding: any grant, contribution, or contract payment that helped pay for the work reduces the SR&ED expenditure base dollar-for-dollar, which makes claim sequencing and documentation a genuine research-administration concern, not just a tax filing.
What SR&ED is and who administers it
SR&ED is a tax incentive program, not a competitive grant program. There is no application deadline in the sense a funding competition has one, no peer review panel, and no fixed pool of money that runs out. Any corporation, individual, trust, or partnership carrying on a business in Canada that performs eligible R&D work in Canada can claim it, by filing the right forms with its annual tax return. The Canada Revenue Agency administers the program — assessing whether claimed work is eligible, reviewing claimed expenditures, and processing the resulting deduction and investment tax credit — while the Department of Finance sets the underlying policy and legislated rates through the Income Tax Act. Because it flows through the tax system rather than a grants competition, SR&ED claims are filed after the work is done, based on the corporation’s own fiscal year, and the credit is realized when the return is assessed.
For research institutions, SR&ED matters less as something universities themselves claim — most universities and other public research institutions are not eligible claimants in the way a private-sector R&D performer is — and more as a program that shapes how industry partners, spin-out companies, and contract research relationships around the institution are structured and funded. A spin-out performing R&D under a sponsored research agreement or licensing agreement with its originating university, or an industry partner co-funding a collaborative project, is very likely to be thinking about SR&ED eligibility as it structures the agreement, and how the institution’s own funding contribution is documented can directly change what the company can claim.
The three eligibility criteria CRA actually applies
CRA does not assess SR&ED eligibility by field, industry, or novelty of the end product. It assesses each individual project, and often each stage of a project, against three tests, applied through what CRA calls its five supporting questions in practice. All three of the following have to be true for work to qualify:
- Scientific or technological uncertainty. There has to be a real technological or scientific obstacle that cannot be resolved using standard practice or publicly available knowledge in the field — something a competent professional in the field, working with existing tools and know-how, genuinely could not have predicted the outcome of in advance. Uncertainty that is purely commercial, financial, aesthetic, or a matter of engineering convenience does not count, even if the underlying work is technically demanding.
- Systematic investigation. The work has to proceed through a recognizable scientific method: a hypothesis or set of hypotheses formulated in advance, a planned test or experiment designed to address the uncertainty, and analysis of the results that feeds into the next iteration. Undirected trial-and-error, or work where the “experiment” is really just building the final product and seeing if it works, does not meet this test on its own — CRA looks for evidence the investigation was planned as an investigation, not just reconstructed after the fact to look like one.
- Technological advancement. The work has to generate new technological knowledge, whether or not it ultimately succeeds. Advancement is assessed against the state of technology at the start of the project, not against what the claimant company itself already knew — and a project that fails to solve the problem it set out to solve can still be eligible, provided the systematic investigation and technological uncertainty are genuinely present, because CRA policy explicitly recognizes that failed experimentation can still generate advancement in knowledge.
Because all three tests apply per project, a single product-development effort commonly contains both eligible and ineligible phases — the underlying uncertainty-resolution work may qualify while the surrounding production scale-up, styling, or commercial rollout does not. CRA’s published SR&ED policy documents, including its guidance on the eligibility of work for SR&ED tax incentives, are the authoritative source for how these tests are applied and are the right place to check current wording before a claim is prepared.
Does the work qualify? A decision aid
This is a starting screen, not a substitute for a real technical review — use it to decide whether a project is worth taking to a qualified SR&ED preparer, not as the final determination.
- Was there a real technological unknown at the start — something a skilled professional in the field could not have solved by looking it up or applying standard techniques? If the honest answer is “we knew how to do this, we just had to do it,” this criterion likely fails.
- Can you point to a documented hypothesis and a test designed to address it — lab notebooks, version-controlled test plans, design-of-experiment records, or equivalent contemporaneous documentation? If the only record is the finished code or product with no trace of the investigation, systematic investigation is hard to demonstrate even if it genuinely happened.
- Did the work generate new technological knowledge, regardless of whether the project ultimately shipped or succeeded commercially? A failed prototype that resolved real uncertainty can still qualify; a successful product built entirely from known techniques cannot.
- Is the uncertainty technological, not commercial? “We don’t know if customers will want this” or “we don’t know if this will be profitable” are business risks, not SR&ED-eligible technological uncertainty.
- Was the work performed in Canada, by or for the claimant, in the tax year being claimed? SR&ED is generally restricted to work performed in Canada, with narrow exceptions.
If a project clears all five, it is a reasonable candidate for a claim; if it clears some but not others, the eligible and ineligible phases usually need to be separated out rather than the whole project claimed or rejected wholesale.
What typically qualifies versus what CRA typically excludes
CRA’s guidance lists categories of work that are excluded from SR&ED by definition, even when they support an otherwise-eligible project, unless the work in that category is itself directed at resolving a technological uncertainty:
| Typically eligible | Typically excluded |
|---|---|
| Experimental development to resolve a specific technological uncertainty in a product or process | Routine engineering or standard design work using known techniques |
| Applied research aimed at a specific practical application | Market research or sales promotion |
| Basic research to advance scientific knowledge with no specific practical application in view | Quality control or routine testing of materials, devices, products, or processes |
| Support work (engineering, design, operations research, math analysis, computer programming, data collection, testing, psychological research) that is commensurate with and directly in support of eligible SR&ED work | Routine data collection unconnected to resolving uncertainty |
| Iterative prototyping and testing where each iteration is driven by a hypothesis about the unresolved uncertainty | Style changes, cosmetic or aesthetic modifications |
| Commercial production and routine commercial exploitation of a proven technology | |
| Prospecting, exploring, or drilling for minerals, petroleum, or natural gas | |
| Social science or humanities research (excluded by statute from SR&ED specifically, though it may be eligible for other funding) |
The most common real-world dispute is not whether R&D happened at all, but where the line falls between eligible experimental development and ineligible routine engineering once a company has already built several generations of a product — CRA’s reviewers specifically probe whether later iterations still involve genuine technological uncertainty or have become incremental, predictable engineering.
Eligible versus ineligible expenditures
SR&ED expenditures fall into defined categories, and the claim method chosen (traditional or proxy, below) changes what can be included:
| Category | Eligible | Not eligible |
|---|---|---|
| Labour | Salaries and wages of employees directly engaged in eligible SR&ED work, prorated to time spent on that work | Time spent on non-SR&ED duties by the same employee; most bonuses/remuneration based on profit |
| Materials | Materials consumed or transformed in the course of SR&ED (prototype builds, test materials) | Materials used in commercial production runs, or capital equipment purchases (post-2014 rule change removed capital expenditures from SR&ED entirely) |
| Contract payments | Payments to arm’s-length Canadian subcontractors for SR&ED performed on the claimant’s behalf, at a reduced inclusion rate | Payments to non-arm’s-length parties beyond cost recovery; payments for work performed outside Canada in most cases |
| Overhead | Traditional method: actual overhead directly related to SR&ED, itemized. Proxy method: a Prescribed Proxy Amount (PPA) of 55% of eligible salary/wage base, claimed in lieu of tracking actual overhead | General administrative and selling costs unrelated to the SR&ED work; capital costs under either method |
| Third-party payments | Certain payments to approved research institutes, universities, or other third parties to have SR&ED carried out on the payer’s behalf, under specific conditions | Payments that are really sponsorships, donations, or unrelated to a specific SR&ED project |
Claimants choose the traditional or proxy method for a given tax year on Form T661; the proxy method’s fixed 55% overhead figure is administratively simpler but is not always the larger number — a company with genuinely high real overhead attributable to SR&ED may recover more under the traditional method’s itemized approach.
Investment tax credit rates and the CCPC distinction
The rate and refundability of the SR&ED investment tax credit depend on corporate structure:
- Canadian-controlled private corporations (CCPCs) below the relevant taxable-capital threshold earn a 35% ITC, fully refundable, on qualifying expenditures up to an annual enhanced-rate expenditure limit. Bill C-15 (Royal Assent March 26, 2026) raised that limit from $3 million to $6 million, effective for tax years beginning after December 15, 2024 — raising the maximum annual refundable credit at the enhanced rate from $1.05 million to $2.1 million. The same legislation raised the taxable-capital thresholds at which the enhanced rate phases out, from $10 million to $15 million (where phase-out begins) and $50 million to $75 million (where it is fully eliminated), and extended eligibility for the enhanced refundable rate to certain Canadian public corporations for the first time.
- Other corporations (non-CCPCs, and CCPC expenditures above the enhanced-rate limit) earn a 15% ITC, which is non-refundable for most corporations — it can only be used to reduce federal tax payable, and unused amounts can be carried back three years or forward twenty.
- Individuals and unincorporated businesses generally earn a 15% credit, of which a portion is refundable, following separate rules from the corporate regime.
The refundable-versus-non-refundable distinction is the single most consequential fact in the program for a small or early-stage company: a refundable credit produces actual cash from CRA even in a year with no taxable income, which is why the CCPC-eligibility and expenditure-limit rules receive so much attention from claimants and advisors. A company that loses CCPC status — through foreign control, going public, or exceeding the taxable-capital ceiling — loses access to the enhanced refundable rate even if its R&D activity is unchanged.
These enhancements were legislated via Bill C-15 in March 2026; CRA’s own SR&ED policy documents and forms were still being updated to reflect them as of this writing. Confirm current thresholds directly against CRA’s SR&ED program pages or with a tax professional before relying on a specific figure for a claim.
Filing: Form T661, deadlines, and what happens after
A SR&ED claim is filed using Form T661, “Scientific Research and Experimental Development (SR&ED) Expenditures Claim,” together with the relevant corporate income tax return (T2) or, for individuals, the T1. Form T661 requires both a technical narrative — describing the scientific or technological uncertainty, the work undertaken to resolve it, and the advancement achieved, for each claimed project — and a financial schedule itemizing the qualifying expenditures. Corporations also file Schedule 31 to calculate the investment tax credit itself.
The filing deadline is strict: eighteen months after the end of the corporation’s tax year in which the expenditures were incurred, with no extensions available. A claim filed even one day late is not eligible for any SR&ED benefit for that year, and CRA has no discretion to accept a late claim. Given that constraint, institutions and companies working with SR&ED-eligible partners should treat the deadline as a hard operational date to track, not a soft target.
After filing, CRA may accept the claim as filed, request more information, or select it for a formal review. Reviews typically involve both a financial reviewer, who examines the claimed expenditures against supporting records (timesheets, payroll records, invoices, project accounting), and a research and technology advisor (RTA), who assesses whether the described work actually meets the three eligibility criteria — often through a site visit or interview with the technical staff who performed the work. Refundable claims are generally processed faster than non-refundable ones under CRA’s service standards, and first-time claimants can request a pre-claim consultation or use CRA’s other pre-claim review services to get an eligibility read before filing.
Government assistance, university collaborations, and the SR&ED expenditure base
This is the interaction that most affects research administrators and technology transfer offices, and it is easy to get wrong: any government or non-government assistance a company receives toward SR&ED work reduces the pool of expenditures that company can claim, dollar-for-dollar, under the Income Tax Act’s assistance rules. “Assistance” is defined broadly and includes federal, provincial, and municipal grants, forgivable loans, and contributions specifically tied to the R&D work — so a company that receives, say, an NRC IRAP contribution or a provincial innovation grant toward the same project has to net that amount out of its SR&ED expenditure pool before calculating the credit. The two programs are not simply additive; funding one project through both a grant and SR&ED does not multiply the total benefit, and claiming SR&ED without correctly reducing for assistance received is a common source of reassessment.
The practical implications for research-administration contexts:
- Sponsored research and collaborative agreements. Where a university or research institute funds part of an industry partner’s R&D directly — cash, in-kind services, or subsidized access to facilities and equipment — that contribution can constitute assistance to the company for SR&ED purposes, reducing what the company can claim. How the funding is structured and documented in the sponsored research agreement or collaboration agreement (a genuine fee-for-service contract versus a grant-like contribution, for instance) affects how it is treated, which is why technology transfer offices and industry-liaison staff negotiating these agreements benefit from involving the company’s tax advisor early, not after the agreement is signed.
- Contract payments between arm’s-length parties. When a company pays a university or research institute to perform SR&ED-eligible work under contract, that payment can be claimed by the paying company as a contract expenditure (at a reduced inclusion rate), while the performing institution generally does not itself claim SR&ED on work done for a fee — the credit attaches to the entity that bore the economic risk of the R&D, not simply the entity that performed the labour. Getting this allocation right, and avoiding both parties claiming the same underlying expenditure, is a documentation issue that should be settled in the contract itself.
- Licensing and spin-out structures. A university spin-out performing SR&ED-eligible development on licensed IP is a distinct corporate claimant from the university, and its own funding history — seed investment, provincial or federal contributions, in-kind institutional support — all factor into its SR&ED assistance calculation. Institutions negotiating the underlying IP and licensing terms with a spin-out, and that continue to provide in-kind support (lab space, equipment access, technical staff time), should be aware that the value of that support can reduce the spin-out’s claimable SR&ED base, even where no cash changes hands.
- Sequencing matters. Because assistance is netted against the SR&ED expenditure pool for the same tax year, and because SR&ED itself has an eighteen-month, no-extension filing deadline, institutions negotiating funding alongside a company’s R&D program should confirm early which fiscal year the assistance will be recognized in and make sure that timeline is compatible with the company’s own SR&ED filing schedule.
None of this is a reason to avoid combining institutional or government funding with SR&ED — most companies with meaningful R&D programs do both — but it is the reason SR&ED cannot be treated as an entirely separate track from grant administration when a university, TTO, or funder is involved in financing the same underlying work.
Frequently asked questions
Is SR&ED a grant?
No. It is a tax incentive delivered through the Income Tax Act and administered by CRA, claimed after the fact on a corporation’s tax return, not a competitive funding program with an application and award process.
Can a university claim SR&ED directly?
Most universities and public research institutions are not typical SR&ED claimants in the way a private-sector performer is; SR&ED mainly becomes relevant to a university through the industry partners, spin-outs, and contract research relationships it works with, and through how sponsored-research funding is structured.
Does a failed R&D project still qualify for SR&ED?
It can. CRA’s eligibility test looks at whether genuine technological uncertainty was investigated systematically, not whether the project succeeded — a failed experiment that resolved real uncertainty and advanced technological knowledge can still be an eligible claim.
What is the difference between the traditional and proxy overhead methods?
The traditional method claims actual overhead costs directly attributable to SR&ED, itemized; the proxy method claims a fixed Prescribed Proxy Amount of 55% of the eligible salary and wage base instead of tracking actual overhead. The choice is made annually on Form T661 and the better option depends on the claimant’s actual overhead structure.
Does receiving a research grant disqualify a company from claiming SR&ED on the same project?
No, but it reduces the claim. Government or non-government assistance tied to the R&D work must be netted out of the SR&ED expenditure pool before the credit is calculated, so the company can still claim SR&ED on the portion of the work it funded itself, but not on the portion covered by the grant or contribution.
This guide summarizes SR&ED program mechanics as of August 2026, drawing on CRA’s published SR&ED program policies and the Department of Finance’s December 2024 SR&ED enhancement announcement, subsequently enacted via Bill C-15 (Royal Assent March 26, 2026). It is general information, not tax advice — SR&ED eligibility determinations are fact-specific and CRA’s own guidance, or a qualified SR&ED tax professional, should be consulted before a claim is prepared or filed.
