R&D tax relief is still available. But the environment around claims has changed.
Founders and finance leads are often led by one question:
"How much can we get back?"
A better question now is: Would our claim stand up if HMRC asked follow-up questions?
Before you file anything, I think there are three questions worth asking.
1. Can You Explain the Technical Uncertainty in Plain English?
Not the commercial goal. Not the business case.
The technical problem.
What were you trying to achieve that wasn't readily deducible by a competent professional? What technical uncertainty did you have to work through? If you can't explain that clearly in writing, the claim is relying on hope, not substance.
A strong claim starts with clarity.
2. Do Your Story, Numbers and Evidence Actually Match?
This is where many claims fall down.
Evidence is not numbers. Time allocations and cost schedules are not "proof". They're just calculations.
The real question is: if you had to justify those costs, what would you point to?
Project notes. Design decisions. Test results. Repo history. Tickets. Meeting notes. Contracts and scopes. A clear link between who did the work and what problem they were solving.
When the narrative says one thing and the costs say another, that's when questions start.
And a spreadsheet on its own is not a claim file.
3. Do You Know Whether You Qualify for ERIS?
A lot of businesses are still operating off outdated assumptions about "which scheme we're in".
For accounting periods beginning on or after 1 April 2024, the regime moved to the merged RDEC scheme, with ERIS for loss-making, R&D-intensive SMEs.
So a very practical question is:
Do you know whether you're ERIS - and can you support that position?
If you don't, you may be modelling the benefit incorrectly, or building a claim on the wrong premise.
R&D relief can still be meaningful for genuine innovation. But the game is no longer "make the biggest claim".
It's "make the claim you can stand behind".
If you're considering an R&D claim and want a clear, straightforward conversation about defensibility and risk (before you submit), we're happy to help.
