- By Michelle Giuricich | Lead Technical Consultant, Catalyst Solutions
R&D tax relief is designed to support UK businesses investing in research and development by providing relief on qualifying R&D expenditure. For companies carrying out qualifying R&D, the benefit can be significant.
However, not all research, innovation or product development will qualify for the relief. HMRC applies a specific definition of R&D for tax purposes, which determines which activities and costs can be included in a claim.
HMRC has also increased its compliance activity following high levels of error and fraud found in claims in recent years. In 2023/24, it carried out around 9,700 R&D compliance checks and identified £441 million that had been incorrectly claimed.
Understanding what qualifies, what does not qualify and which costs can be included is an important first step before preparing a claim.
In the CIRD Manual, HMRC provides guidance on the meaning of R&D for tax purposes. It is defined as work that seeks to achieve an advance in science or technology through the resolution of scientific or technological uncertainty.
There are three key requirements.
The work must seek to increase overall knowledge or capability within a field of science or technology. Something being new to your company is not enough if the underlying solution is already known or readily available within the wider field. An advance does not have to mean inventing something completely new. It can include an appreciable improvement to an existing product, process, material, device or service.
There must be uncertainty about whether the desired outcome is scientifically possible or technologically feasible, or about how it can be achieved in practice. A project being difficult, expensive or time-consuming does not automatically make it R&D. If the solution can readily be found using existing knowledge or established methods, the work would not normally qualify.
The uncertainty must extend beyond the knowledge of your own team. If a suitably experienced professional with current knowledge of the relevant field could readily work out the solution using existing knowledge, the work would not normally qualify.
Most R&D projects in the commercial world begin with a wider business objective.
You may be developing a new product, improving the performance of existing software, reducing energy or material use, automating a complex process, improving the accuracy of a system, increasing production capacity or finding a better way to achieve an existing outcome.
It is important to note that qualifying R&D is not based on the outcome of the product being developed, but rather on the science or technology behind it. Start with the business objective, then consider what prevented you from achieving it using the scientific knowledge, technology or methods already available.
The commercial objective and the qualifying R&D are not necessarily the same thing. A business may want to reduce manufacturing waste, for example, but the potential R&D may sit within the technological problem its engineers need to solve to make that possible.
Similarly, a software company may want to process customer data faster. The potential R&D would not be the new platform itself, but the work required to overcome a technological limitation that existing approaches could not readily solve.
This is often the easiest way to start identifying where the R&D sits within a wider project.
To identify where qualifying R&D may sit within your project, ask yourself:
Be specific about the product, process, system or capability you were developing or improving.
What was the baseline knowledge available at the time? Was there already technology, published knowledge or an established method available that could achieve what you needed?
Think about the particular scientific or technological problem, rather than whether the overall project was difficult.
Did you need to develop, design, test or experiment with different approaches before reaching the required outcome?
Separate the work aimed at overcoming the scientific or technological uncertainty from routine development, implementation and commercial activity.
A project does not need to succeed for the work to qualify. A failed attempt can still involve qualifying R&D where the project was seeking an advance and the team undertook work to resolve a genuine scientific or technological uncertainty.
The key question is whether a competent professional could determine the outcome or method using existing knowledge, or whether further investigative or experimental work was needed to establish it.
The qualifying criteria can look very different across industries. While the examples below focus on manufacturing, software and agriculture, businesses across a wide range of sectors can undertake qualifying R&D.
A manufacturing client approached Catalyst Solutions after increasing the speed of a production line used to mould plastic components. Our technical team’s first question was whether this could have been achieved simply by installing faster machinery and running it on the supplier’s standard settings. If so, the work would be unlikely to qualify, as the solution was already established.
In discussion with the client’s competent professional, it became clear that this was not the case. Running the line at higher speeds caused the material to deform, and the existing machinery could not maintain the required temperature and pressure tolerances. The engineering team had to test different materials, tooling configurations and process conditions to find a workable solution. We then worked with the client to develop a technical narrative setting out the technological uncertainty they faced and the work undertaken to resolve it, which may qualify as R&D.
A software client came to us while developing a logistics platform that tracks thousands of deliveries and provides live updates. Our technical team first explored whether the platform had been built using established databases, APIs and development frameworks, since a system being new to the business does not, on its own, make the work qualifying.
The client’s development lead explained that existing architecture could not process the required volume of live location data quickly enough, and the team could not readily determine how to achieve the required performance. They had to design and test different technological approaches before arriving at one that worked. Together, we documented that technological limitation and the work undertaken to overcome it, which may qualify as R&D.
An agricultural technology client asked us to review its work on an irrigation system that automatically adjusts water and nutrient delivery according to changing soil and environmental conditions. We first considered whether the system simply used established sensors, software and irrigation methods in a standard configuration, adapted for a particular farm. If that were the case, the work would be unlikely to qualify.
Speaking with the client’s technical team, we learned that existing systems could not respond reliably to the level of variation required. The team had to test different combinations of sensors, control methods and delivery mechanisms to achieve a dependable result. The client’s commercial goal was to improve crop yields, but in developing the technical narrative we focused on the technological uncertainty they had to resolve to reach that goal, as this is where the potential R&D lies.
Once the qualifying R&D activities have been identified, the next step is determining which costs relate to that work.
It is important to note that only costs relating to qualifying R&D activities should be included in the claim.
Qualifying expenditure can include:
Any costs relating to work being done after the technological challenge is resolved or the scientific or technological advance is achieved (eg consumer testing) would be considered costs outside of the R&D boundary. Similarly, work done prior to a project being established (eg certain feasibility work) may also be considered non-qualifying.
Where a cost relates partly to qualifying R&D and partly to other work, only the appropriate R&D proportion should be included. For example, if a developer spends 40% of their time working on a qualifying technological problem and the remaining 60% on routine functionality, only the relevant proportion of their staffing cost should form part of the claim.
Certain indirect activities that support the R&D can also qualify, such as specific administration, maintenance, training or feasibility work connected to the R&D project.
R&D tax relief can provide significant support where the underlying work meets HMRC’s qualifying criteria.
The key is to identify the scientific or technological work within the wider project, separate it from routine or commercial activity, and make sure the costs included in the claim relate directly to that qualifying R&D.
If you are unsure where that boundary sits within your own projects, a detailed review of the work undertaken can help determine which activities and costs may qualify. Learn more about Catalyst Solutions‘ UK R&D tax relief services.
ABOUT THE AUTHOR
Michelle Giuricich is the Lead Technical Consultant at Catalyst Solutions. She works across industries, overseeing the technical assessment on every R&D project the firm takes on. Her background as a chemical engineer at Sasol shapes how she reads complex technical work and what she looks for when assessing whether a project meets the uncertainty test.
© 2026 Catalyst Solutions – Alle Rechte vorbehalten | Allgemeine Geschäftsbedingungen
