With HMRC’s applying heavy scrutiny to R&D tax credit claims, many companies that previously made their claims without much difficulty may be giving their claim a second look now. You may assume that if your R&D project relies on existing software, equipment or methodologies, you don’t meet the standard for an “advance in science or technology”.
That assumption is wrong, but it is not quite as straightforward as "yes, everything counts." The real answer depends on what you are doing with those tools, and whether your work meets HMRC's definition of R&D.
Key takeaways
- Claiming R&D tax credits does not require you to avoid existing tools. What matters is whether your project seeks a genuine advance in science or technology.
- Customisation and minor improvements to off-the-shelf tools do not qualify. Substantial advances that push genuinely beyond the current state of the art may well do.
- Using standard tools, libraries, and methodologies within your R&D project is permitted and does not disqualify the work.
- The costs of software licences and other tools used in qualifying R&D can be included in your claim, apportioned to the R&D element.
How does HMRC define R&D?
Before we get into off-the-shelf tools specifically, it is worth revisiting the definition. HMRC's definition of R&D for tax purposes centres on one question: are you seeking an advance in science or technology?
That advance must be in the overall knowledge or capability in a field, not just your own understanding of it. An advance in science or technology could include an “appreciable improvement”, which relates to existing technologies.
To achieve an advance, your project must overcome "scientific or technological uncertainty", or a genuine doubt about whether something is achievable that a competent professional in the field could not readily resolve.
What is an "appreciable improvement"?
One of the most important phrases in the HMRC’s guidelines is appreciable improvement. A qualifying advance does not have to be revolutionary, but it does have to be substantive. Routine or incremental improvements (the kind of thing a developer or engineer would do in the normal course of their work) do not qualify.
HMRC specifies that an appreciable improvement should “generally be acknowledged by a competent professional working in the field as a genuine and non-trivial improvement”. This is why talking to your technical team is so important when determining if your project qualifies.
This is where off-the-shelf tool questions tend to land. If you take an existing tool and make a minor configuration change, add a standard integration, or customise it within its designed parameters, that is not an appreciable improvement on the state of the art. The capability of the field has not moved forward.
But, if you are pushing a tool beyond what it was designed to do, resolving a genuine uncertainty about whether it can perform in a new context, or developing something genuinely novel on top of it, you may well have a qualifying project.
How does this work in practice?
Here is how the question plays out across some common situations we see.
Improving upon an off-the-shelf tool
This can qualify, but the bar is set at a genuine improvement on the state of the art. If you are working to extend a tool's capabilities in a direction that was previously uncertain, technically difficult, or unproven, and a competent professional in your field would not have known at the outset how to do it, that work may qualify. Customisation and minor configuration changes do not clear the bar.
Worth noting: the fact that the work was difficult or time-consuming for your team is not, on its own, sufficient.
Using off-the-shelf tools within your R&D project
This is the most common situation, and in practice the answer is generally yes. If your R&D project is seeking an advance in its own right, you can use standard tools, code libraries, SDKs, and testing methodologies to carry out that work without disqualifying the claim.
HMRC does not require you to build everything from scratch. Using an existing testing framework to validate results, or an established SDK as the foundation for novel development, is entirely within scope.
Claiming the costs of those tools
If you use a software licence, a cloud computing platform, or a piece of equipment as part of your qualifying R&D, a proportion of those costs can be included in your claim. The key word is "proportion": only the element of usage that relates to the R&D activity is eligible. If a licence serves both R&D and non-R&D purposes, you will need to apportion it fairly.
Qualifying and non-qualifying examples by sector
The distinction between qualifying and non-qualifying work is clearer with concrete examples. In each case, the tool is identical; what changes is the nature of the work being done with it.
Software
Company A, a 30-person software business, is building a machine learning model on top of an open-source framework. The team is resolving genuine uncertainty about whether the approach can achieve accurate predictions at the required speed, given the volume and structure of the data involved. The framework is off-the-shelf; the advance being sought is not. This qualifies.
Company B uses the same framework to build a new feature that falls within the known capabilities of that framework. No uncertainty, no advance; this does not qualify.
Engineering
Company A uses existing simulation software to test a new materials configuration that has not been validated before. The software is standard but the unknowns being resolved are not. This qualifies.
Company B uses the same software to run simulations of an established process with a slight adjustment to confirm the quality is still in tolerance. That is standard professional practice and does not qualify.
Manufacturing
Company A adapts an off-the-shelf automated assembly system to handle a new composite material with unpredictable mechanical properties, where it was genuinely uncertain whether reliable performance was achievable at the required tolerance. The adaptation required significant technical investigation over several months. This qualifies.
Company B installs an off-the-shelf system as intended, for a known material, within the system's stated operating parameters. They adjust their parameters to meet their output requirements. This does not qualify.
Pharma
Company A uses an established testing methodology to investigate the binding behaviour of a new compound, where the outcome was genuinely uncertain at the outset. The test is standard but the scientific uncertainty being resolved is genuine. This qualifies.
Company B uses the same testing methodology to run standard quality control tests on an existing compound as part of routine production. This does not qualify.
Making a judgement on your R&D claim
HMRC has significantly increased the volume and intensity of compliance checks on R&D claims in recent years. Claims involving off-the-shelf tools are not inherently higher risk, but any claim that lacks clear evidence of scientific or technological uncertainty, or that describes work resembling standard professional practice, is more likely to attract attention.
Deciding whether your use of tools qualifies, and how to apportion costs fairly, are exactly the kinds of questions you need to get right. Compliance is critical for R&D tax claims, but just as important to your business is claiming every eligible project.
Tax Cloud's team of R&D experts reviews every project submission before it goes to HMRC. We work through the detail of your work with you, so you are not left guessing about which activities qualify or how to treat the costs involved. If you would like to talk through a specific project, get in touch and we will walk you through it.
