Does your business qualify for R&D tax relief? A plain-English eligibility check

Research and development tax relief is one of the most misunderstood parts of the UK corporate tax system. Some companies assume anything described internally as “innovation” qualifies. Others assume relief is only for laboratories and universities, and never look at work that might genuinely fall within the definition. Neither starting point is a reliable guide to the statutory test.

This article is a preliminary eligibility check. It is written for directors and finance leads who want to work out whether a project looks worth investigating properly, before spending time gathering costs or engaging anyone to prepare a claim. It is indicative only. Nothing here decides whether a claim is valid: that depends on the specific technical facts, the judgement of a competent professional in the relevant field, and the evidence you can produce.

Three separate tests, not one

The single most useful thing to understand at the outset is that eligibility is not one question. There are three distinct tests, and a project can pass one and fail another:

1. Company eligibility — whether the entity making the claim is within the charge to UK Corporation Tax and meets the conditions attaching to the relief it is claiming under.
2. Project eligibility — whether the activity meets the statutory definition of research and development for tax purposes, which turns on an advance in science or technology and the resolution of scientific or technological uncertainty.
3. Cost eligibility — whether the particular expenditure you want to include falls within the categories of qualifying cost, and is attributable to the qualifying activity.

Tests two and three are where the detail usually sits. A company can be eligible, be doing something genuinely difficult, and still find that some of the cost it hoped to include sits outside the qualifying categories or belongs to another party. Treating the three as separate questions from the beginning keeps expectations realistic.

What the relief is intended for

The relief exists to support work that seeks to advance overall knowledge or capability in a field of science or technology, not work that is simply new to the company doing it. That distinction is doing an enormous amount of work in practice. Buying, configuring and deploying capability that already exists in the field — however difficult, expensive or commercially significant — is not, on its own, an advance in the sense the legislation means.

The meaning of research and development for tax purposes is set out in guidelines issued by the Department for Science, Innovation and Technology, which HMRC applies. Those guidelines, rather than any commercial description of the relief, are the reference point worth reading if you are trying to form your own preliminary view.

Advance in science or technology

An advance means an increase in overall knowledge or capability in a field of science or technology, not an increase in the company’s own knowledge. The test looks outward, at the state of the field, rather than inward at the team’s learning curve.

Practically, that means asking: if a knowledgeable professional in this field were shown what we set out to achieve, would they say the way to do it was already known and documented? If the answer is yes, the project is applying existing capability, even if doing so was hard work. If the answer is that competent professionals in the field would not readily know whether it could be done, or how, the project may be seeking an advance.

An advance can also exist where something already achieved by others is genuinely not available in the public domain and cannot reasonably be deduced by a competent professional. That is a narrow route and depends heavily on evidence about what was and was not knowable at the time.

Scientific or technological uncertainty

Uncertainty in this context is technological, not commercial. It exists where knowledge of whether something is scientifically possible or technologically feasible, or how to achieve it in practice, is not readily available or deducible by a competent professional working in the field.

Several things that feel uncertain to a business are not technological uncertainty:

• Whether customers will buy the product, or whether the market will accept a new pricing model.
• Whether the project can be delivered within the budget or the deadline.
• Whether the team has the skills, or whether a supplier will deliver on time.
• Whether an approach that is documented and understood in the field will work in this particular company’s environment, where the method itself is established.

Uncertainty also has to be resolved through a systematic attempt — a process of investigation, iteration and testing directed at overcoming it — rather than encountered incidentally.

The role of the competent professional

Whether an advance and an uncertainty exist is judged from the perspective of a competent professional working in the relevant field. That is a person with the knowledge, qualifications and practical experience to know what is already established, what is genuinely difficult and where the boundary of current capability sits.

This matters for two reasons. First, the analysis is technical rather than financial: it must come from someone with relevant scientific or technological knowledge and experience. An accountant or tax adviser may coordinate the claim, but should not substitute for technical competence they do not possess. Second, the person forming the view should be able to explain, in their own words, what was not known, why existing approaches were inadequate and what was attempted. If nobody within the company or closely involved with the project can do that, the project is probably not ready to be examined further.

Qualifying and non-qualifying activities

Where a qualifying project exists, activities that directly contribute to resolving the uncertainty generally fall within it, and certain supporting activities may as well. Activities outside the project boundary do not, even where they are essential to the commercial outcome.

The project also has a start and an end for these purposes. It generally begins when work to resolve the uncertainty starts, and ends when that uncertainty is resolved or work to resolve it ceases — not when the product ships.

What “innovation” does not mean here

A recurring problem is language. Boards, investors and grant applications use “innovation”, “novel”, “first of its kind” and “breakthrough” in a commercial sense. The statutory test does not follow that usage. None of the following automatically satisfies it:

• Being first to market with a product or business model.
• Commercial novelty, or being new to the company or its sector.
• Commercial uncertainty about whether an investment will pay off.
• Product improvement that applies established techniques.
• Winning an innovation award or an innovation-framed grant.

Software deserves a specific note, because blanket statements are made in both directions. It is not correct that integration, interface or framework work always fails the test, and it is not correct that software development always passes it. The reliable formulation is narrower: routine application of established methods, tools and patterns generally does not demonstrate an advance in the field, while the specific technical facts — what was attempted, what was not knowable and why — determine the result in any given case.

Three fictional examples

The following examples are invented for illustration. They are not case studies, not clients and not precedents. In each case the actual outcome would depend on the facts, on the competent professional’s analysis and on the evidence available.

Example one: potentially qualifying

A fictional manufacturer of industrial sensors needs its devices to remain accurate in an environment with vibration and temperature swings well outside the operating range for which existing sensing approaches are documented. Its engineers try several established compensation methods, all of which fail in testing, and then develop and iterate on an approach whose feasibility they cannot determine in advance. Whether this qualifies would depend on whether a competent professional agrees the required capability was not readily deducible, and on what the testing records show.

Example two: likely not qualifying

A fictional wholesaler connects its e-commerce platform to a widely used accounting package and a third-party warehouse system, using documented interfaces from each vendor. The project takes months and involves substantial data-cleansing effort. The work is commercially valuable and demanding, but the methods are established and documented, so on these facts it is unlikely to show an advance in the field. As always, the outcome depends on the specific technical facts, the competent professional’s view and the evidence.

Example three: genuinely borderline

A fictional software business needs a query workload to return results far faster than its current architecture allows. Standard optimisation techniques are applied first and get part of the way. The team then attempts a restructuring whose viability at the required scale it says is not documented anywhere it can find. Whether the second phase involved technological uncertainty, or was skilled application of known techniques, is exactly the kind of question that needs a competent professional’s reasoned view and contemporaneous evidence rather than a general article.

Evidence and records to keep as you go

Evidence assembled after the fact is weaker and more expensive to produce than evidence captured while the work happens. If you think a project may be within scope, start recording contemporaneously:

• A short written statement of the intended advance, and what was not known at the outset, in the competent professional’s own words.
• Which existing approaches were considered or attempted, and why they were inadequate.
• Design notes, test plans, results and records of failed attempts.
• Who worked on the project, in what role, and a defensible basis for the proportion of their time spent on qualifying activity.
• Dates marking when work to resolve the uncertainty began and when it ended.
• Supporting cost records tied to the project rather than to the department or the financial year.

The claim process at a high level

A claim is made through the company tax return for the relevant accounting period. Two administrative steps sit alongside it, and both are worth understanding early because they have timing consequences.

Claim notification

Some companies must notify HMRC in advance that they intend to claim, using a claim notification form. The notification period is determined by the company’s period of account and generally ends six months after the end of that period of account. Whether notification is required at all depends on the company’s claim history, the exceptions that may apply and the relevant periods. This is a deadline-driven point where getting it wrong can remove the ability to claim, so check your own position against HMRC’s claim notification guidance rather than relying on a general summary.

Additional Information Form

Detailed information supporting the claim must be submitted to HMRC before the claim itself. It covers matters such as project descriptions written against the statutory test, the qualifying costs, and details of the people responsible. The requirements are specific and are set out in HMRC’s guidance on submitting detailed information.

Areas that need specialist review

Several areas are technically involved, fact-specific and frequently misjudged. None of them should be resolved from a general article, and each is a reason to bring in specialist input before assuming a cost is claimable:

• Contracted-out research and development — where work is done for a customer, entitlement depends on the statutory conditions and the specific facts. It cannot be determined simply from invoicing, contractual labels or any single risk factor, and is worth reviewing with a specialist.
• Subcontractor costs — at a high level, the treatment depends on the relief being claimed and on the relationship between the parties, and warrants specialist review.
• Externally provided workers — arrangements through staff providers have their own conditions and are best reviewed with a specialist.
• Overseas expenditure — restrictions apply and the exceptions are narrow and fact-dependent.
• Grants, subsidies and other funding — funding arrangements may interact differently with scheme choice, with earlier accounting periods and with the company’s wider circumstances, so current official guidance or specialist advice is needed.
• Group and connected-company arrangements — which entity incurs the cost, employs the people and bears the risk matters.
• Overlap with creative-sector reliefs — expenditure supported under a creative-industry regime cannot simply be counted twice. If your work sits in film, television, animation or video games, see our overview of creative industry tax reliefs and, for games specifically, the Video Games Expenditure Credit.

When specialist input is appropriate

A preliminary check of the kind set out above is useful for deciding whether to look further. It is not a substitute for a proper assessment. Specialist input is appropriate where the project appears to involve genuine technological uncertainty, where any of the areas listed above is in play, where the amounts involved are material, or where the company has not claimed before and the notification position needs to be established.

Singletree is a general practice, not a specialist research and development claims firm. Where a project looks worth investigating, we will say so, help you think about what evidence exists, and where appropriate suggest involving a specialist who works in this area. We do not prepare or promote claims on the basis of expected outcomes. Founder-stage companies weighing this alongside other decisions may also find our startup accounting support a more useful starting point, and we summarise wider changes on our tax updates page.

If you would like a second view on whether a project appears worth investigating, you can book a meeting to talk it through.

Sources and further reading

The following official sources were checked on 17 August 2026. Rules change, and you should check the current version of each before acting.

• Corporation Tax: Research and Development (R&D) tax relief — GOV.UK
• R&D tax relief: the merged scheme and enhanced R&D intensive support — GOV.UK
• Tell HMRC that you’re planning to claim R&D tax relief — GOV.UK
• Submit detailed information before you claim R&D tax relief — GOV.UK
• Guidelines on the meaning of research and development for tax purposes — GOV.UK