When copying can qualify as R&D for tax purposes
Copying or adapting something that already exists can qualify when your work aims to achieve an advance in science or technology and you face uncertainty that competent professionals cannot readily resolve using existing knowledge.
The key test: uncertainty + advance
To qualify, you need to show:
- What advance you were trying to achieve (in scientific or technological terms), and
- What uncertainties stopped you getting there at the outset, and;
- How you tried to resolve them through systematic work (design, prototyping, trials, testing, analysis)
'New to your business' is not enough
If the method is already known in the field and a competent professional could readily do it, HMRC is unlikely to accept it as R&D, even if it’s new to your team.
Common scenarios where copying is more likely to qualify
These are patterns that often involve real uncertainty (the facts will always depend on your specifics).
Reverse engineering where the 'how' isn’t deducible
You may be replicating an outcome, but key details aren’t publicly available or aren’t readily deducible by a competent professional, so you still face genuine uncertainty.
Replicating performance under new constraints
Copying becomes R&D-like when constraints force new technical solutions, for example:
- Different materials or supply chain substitutes
- Tighter tolerances
- New operating environments (temperature, corrosion, sterilisation, load)
- Scaling from prototype to reliable production
Achieving an appreciable technological improvement
Improving an existing product or process can qualify where the improvement is non-trivial in technological terms (not just 'better looking' or 'easier to sell').
When copying does not qualify
Copying is unlikely to qualify if it is mainly:
- Routine replication using standard methods
- Incremental tuning, debugging, or normal optimisation
- Quality control or compliance testing without new uncertainty
- Commercial uncertainty (pricing, customer preference, supplier lead times) rather than technological uncertainty
An useful rule of thumb is: If a competent professional could have delivered the result using established techniques without a meaningful technical unknown, HMRC is likely to see it as non-R&D.
How to explain it in your claim and what HMRC needs to see
Your narrative should make it easy to follow the logic from baseline → uncertainty → work → outcome.
Start with the baseline
- What already existed (product/process/knowledge)
- What you could and couldn’t readily know from public sources
- Why existing information didn’t solve your problem.
Describe the uncertainties clearly
Be specific. 'We weren’t sure if it would work' is weak. A better way to frame would be: 'We couldn’t achieve X performance at Y tolerance without Z failure mode, and the solution wasn’t readily deducible.'
Show the systematic work
Outline iterations and learning:
- Prototypes and trials
- Modelling and simulation
- Test design and results
- Failed approaches and what you learned
Evidence to keep for “copying” projects
The strongest claims don’t just tell a story, they can back it up.
Practical evidence examples
- Test plans, lab reports, trial outcomes
- Design logs, CAD/version history, change notes
- Tickets/commits (software) showing iterative problem-solving
- Meeting notes where technical decisions were made
- Records showing who your competent professionals were and what they concluded.
Does the merged scheme change this?
No. From 1 April 2024, most claims fall under the merged R&D tax relief scheme, with ERIS available for certain R&D-intensive loss-making SMEs, but the definition of R&D is still based on the same HMRC guidelines.
.svg)
