What Qualifies as R&D for Tax Purposes?

Key facts

R&D must seek an advance in overall science or technology — not just be new to your company.
The project must face scientific or technological uncertainty that a competent professional could not readily resolve.
Work must follow a systematic approach — trial and error alone is not enough.
The BIS Guidelines (published by the Department for Science, Innovation & Technology) define qualifying activities.
Both successful and failed projects can qualify, provided the uncertainty was genuine.

The BIS Guidelines Test

The UK definition of R&D for tax purposes is set out in the BIS Guidelines (originally published by the Department for Business, Innovation & Skills, now maintained by the Department for Science, Innovation & Technology). HMRC uses these guidelines to determine whether your project qualifies.^[2]

To qualify, your project must meet all three of the following conditions:

Condition	What It Means
Advance	The project seeks an advance in overall knowledge or capability in a field of science or technology.
Uncertainty	There is scientific or technological uncertainty — a competent professional in the field cannot readily determine how to achieve the advance.
Systematic approach	The work follows a systematic, investigative method to resolve the uncertainty (hypothesis, testing, analysis).

Key distinction: The advance must be in science or technology, not in business processes, marketing, the arts, or social sciences. Commercial innovation alone does not qualify unless it is driven by a genuine technological challenge.

What Counts as an Advance?

An “advance in science or technology” means achieving something that extends the overall knowledge or capability in the field — not merely extending your own company’s knowledge.^[2]

Advances can include:

Creating a new product, process, material, device, or service through scientific or technological advances
Making an appreciable improvement to an existing product or process through resolving scientific or technological uncertainty
Using science or technology to duplicate the effect of an existing product or process in a new or appreciably improved way
Developing new scientific knowledge or capability

The advance does not need to be a ground-breaking discovery. Incremental improvements that resolve genuine uncertainty can qualify, provided they go beyond the application of existing knowledge.

Scientific or Technological Uncertainty

This is the most critical test. Uncertainty exists when a competent professional working in the field cannot readily determine:^[3]

Whether something is scientifically or technologically achievable at all
How to achieve it in practice
What the outcome will be when attempting to achieve it

If the solution is already publicly available or can be easily worked out by a competent professional using standard techniques, there is no qualifying uncertainty.

Tip: When writing up your R&D claim, clearly articulate what the uncertainty was at the outset. Ask yourself: “Could a competent professional have known the answer before we started?” If the answer is yes, the work likely does not qualify.

What Does Not Qualify?

The following activities are explicitly excluded from qualifying R&D:^[2]

Work in the arts, humanities, or social sciences (including economics)
Routine testing, quality control, or data collection
Cosmetic or stylistic changes that do not involve science or technology
Commercially innovative work that uses established technology in standard ways
Market research, business planning, or feasibility studies (unless they involve resolving scientific uncertainty)
Training or professional development

Activities That Support Qualifying R&D

Certain activities that are not R&D in themselves can still qualify if they are carried out as part of, or in support of, a qualifying R&D project. These “qualifying indirect activities” include:^[4]

Scientific or technological planning related to the R&D
Design, testing, and analysis directly related to resolving the uncertainty
Mathematical analysis to support the R&D
Clerical and administrative work essential to the R&D project (such as recording findings)
Training staff specifically to carry out the R&D

The Competent Professional

A “competent professional” is a person with the relevant qualifications, experience, and knowledge in the field of science or technology in question. This is the benchmark against which uncertainty is measured.^[3]

For your claim to succeed, you should identify a competent professional (or team) who can confirm that the uncertainty was genuine. This person does not need to be an external consultant — they can be an employee, such as a senior developer, engineer, or scientist within your organisation.

Frequently Asked Questions

Does my project need to be completely novel to qualify?

No. Your project does not need to be a world first. It must seek an advance in science or technology that is not already publicly available or readily deducible by a competent professional in the field. If the knowledge needed to solve your problem was not publicly available, your project may still qualify even if others have solved the same problem independently.

Can routine product development qualify as R&D?

Not usually. Routine development using established techniques, standard engineering, or off-the-shelf solutions does not qualify. However, if your product development hits a genuine technological uncertainty — for example, making a material perform beyond its known limits — the work to resolve that uncertainty can qualify.

Do software projects qualify for R&D tax relief?

Yes, software projects can qualify if they seek to achieve an advance in computer science or information technology. This might include developing new algorithms, creating systems that handle data in ways not previously achievable, or solving complex integration challenges. Straightforward app development using known techniques typically does not qualify.

What if my R&D project failed?

Failed projects can still qualify. R&D tax relief is about the attempt to resolve uncertainty, not about achieving a successful outcome. If you undertook systematic work to try to advance science or technology but ultimately could not overcome the uncertainty, the costs of that work are still eligible.

Sources

Corporation Tax: Research and Development (R&D) relief — GOV.UK
Guidelines on the meaning of research and development for tax purposes — GOV.UK / DSIT
CIRD81900 – R&D tax relief: conditions to be met – the R&D — HMRC
CIRD81300 – R&D for tax purposes: the BIS guidelines — HMRC