We were told we probably qualify for R&D relief, but nobody will tell us what happens if the claim is wrong.
R&D relief is worth having and it is now heavily scrutinised. The claims that fail are rarely the ones without merit. They are the ones without a contemporaneous record of what was actually attempted and why it was uncertain. We build the record as the work happens, and coordinate the specialists who file. The change in the last few years has not been to the rules so much as to the standard of evidence expected behind them, and a claim assembled from timesheets and recollection at the year end now reads exactly like what it is.
Dinmore Bell builds and holds the evidence base behind an R&D relief claim: the technological uncertainty, the competent professional, the work done and the costs attributed, captured while the work happens. Accountants or specialist advisers prepare and submit the claim under its coordination, and Dinmore Bell owns the response if an enquiry opens.
How we own it
What actually qualifies
An honest assessment before any work starts, against the current guidance rather than the one that applied three years ago. If the answer is that a project does not qualify, you get that answer for free and early. The test that decides most claims is not whether the work was difficult or new to the company. It is whether there was an advance in a field of science or technology, and whether a competent professional in that field would have regarded the outcome as genuinely uncertain at the outset. Plenty of valuable and demanding engineering work fails that test because the uncertainty was commercial or one of integration effort rather than technological. Plenty of unglamorous work passes it because a known approach was being pushed beyond its established limits. Where the line falls matters most for software, which is where the largest share of contested claims sits. Building a system that has not been built in your business before is not an advance if the techniques are established and documented. Resolving a performance, scale or interoperability problem that the available literature and tooling do not answer often is. The distinction is defensible either way, but only if somebody wrote down which side of it the project was on, at the time, and why.
The evidence, built as you go
The competent professional, the technological uncertainty, the work done to resolve it, and the costs attributed to it, captured while the project runs rather than reconstructed a year later from timesheets and memory. In practice that is five things. A named competent professional for each project, with a short record of why that person holds that status, which is a matter of experience and standing in the field rather than a job title. A written statement of the uncertainty at the point it was identified, before the answer was known. A log of the approaches attempted, including the ones that failed, because failed attempts are the strongest evidence that the outcome was genuinely uncertain. A basis for apportioning staff time that reflects what people actually did rather than a flat percentage applied to a department. And a clear boundary showing where the qualifying work stopped and ordinary development resumed. We set that up so it lives in the systems the team already uses, because a process that requires engineers to fill in a separate form produces nothing by month three. Usually it means a short project record created when a project is opened, a standing prompt in the existing sprint or stage-gate routine, and a quarterly review that catches drift while the people involved still remember the detail.
The claim, and what happens after it
We coordinate the accountants and specialist advisers who prepare and file, hold the timetable, and own the response if HMRC opens an enquiry. An enquiry answered properly in month one rarely becomes anything else. The division of labour matters. The computation, the return and the additional information requirements are your accountant's or a specialist adviser's work, and Dinmore Bell does not perform them. What sits between the technical team and the filing is the part usually left to chance: making sure the narrative submitted matches the record that exists, that the costs claimed reconcile to the apportionment basis actually used, and that the person who will have to defend the claim has read it before it goes. When an enquiry opens, the first letter sets the tone for everything after it. A response that answers the questions asked, attaches the contemporaneous record and offers a call with the competent professional is a very different document from one that restates the original narrative in longer form. We hold the correspondence, prepare the people who will be spoken to, and keep the commercial consequences in view, because an enquiry running into a funding round or a sale becomes a disclosure item whether or not it is resolved.
Subcontractors, grants and the contracts that decide the claim
A great deal of R&D relief turns on contract terms rather than on technology, and this is where the commercial and the technical meet. Where work is subcontracted, who may claim depends on who bore the financial risk, who decided what was to be attempted, and who owns the result. Those three things are set by the subcontract, and a development agreement drafted without the relief in mind frequently gives them away. Where a customer is paying for the work, the question is whether the project is contracted R&D for that customer, which changes the position again. Grant funding interacts with relief in ways that are easy to get wrong and expensive to unwind, because subsidised expenditure and the project boundary have to be handled deliberately rather than discovered afterwards. And where intellectual property in the output sits with a contractor, or with a university partner, or in a group company other than the one incurring the cost, the claim and the ownership can end up in different places. Dinmore Bell reads and writes those agreements: development contracts, collaboration and consortium agreements, grant funding agreements and intra-group arrangements. The aim is narrow and practical, which is that the contractual position supports the claim the business intends to make, decided before the work starts rather than after.
How the function runs
A retained function on a fixed monthly fee, with the R&D evidence base held as part of the wider contract and IP register rather than as a standalone project. That means the qualifying projects are identified when they open, the competent professional is named at that point, the record is kept as the work runs, the contracts behind the work are read before they are signed, and the filing timetable is held against the accountant's diary rather than yours. If an enquiry arrives, the file already exists. The tax computation, the submission and any regulated tax advice stay with your accountants or a specialist adviser. Dinmore Bell coordinates them, holds the evidence, and owns the correspondence.
Where specialists come in
We do not give regulated tax advice and we do not file your return. The technical tax treatment and the claim itself are prepared by your accountants or by instructed specialist advisers, whom we coordinate and hold to the timetable. Our work is the evidence, the project record and the management of the process.
Sectors we run this for
Related insights
All insights →What a fully loaded General Counsel actually costs
The comparison is not our fee against an hourly rate. It is our fee against the hire you would otherwise make.
Where nine per cent of your revenue goes
Contract value rarely leaks in a single expensive event. It leaks quietly, through renewal dates nobody owns.
- What does HMRC actually want to see behind an R&D claim?
- A named competent professional, a statement of the technological uncertainty identified before the outcome was known, a record of the approaches attempted including those that failed, and a defensible basis for the costs apportioned to the project. The strongest claims show that record being created as the work happened rather than assembled afterwards.
- Does software development qualify for R&D relief?
- Some does. Building something new to your business using established techniques generally does not, because there is no advance in the field. Resolving a scale, performance or interoperability problem that the available literature and tooling do not answer often does. The deciding factor is whether the uncertainty was technological, and whether that was recorded at the time.
- Who can claim when the work is subcontracted?
- It depends on who bore the financial risk, who decided what was to be attempted, and who owns the result, and all three are set by the subcontract. A development agreement written without the relief in mind often gives those away, which is why the contract is worth reading before the work starts rather than at the year end.
- Can we still claim if the project received grant funding?
- Frequently yes, but the interaction between subsidised expenditure and the project boundary has to be handled deliberately. The costs, the project definition and the funding agreement need to be considered together at the outset, because unwinding the position after a claim has been filed is considerably harder than planning it.
- What happens if HMRC opens an enquiry?
- Dinmore Bell owns the correspondence, assembles the contemporaneous record, prepares the competent professional for any discussion, and keeps your accountant's technical work aligned with the answers given. An enquiry answered in the first month with the actual record attached rarely escalates; one answered by restating the original narrative usually does.
- Does Dinmore Bell prepare and file the claim?
- No. The computation, the return and the additional information requirements are prepared and submitted by your accountants or a specialist R&D adviser. Dinmore Bell builds and holds the evidence base, reads the contracts the claim depends on, coordinates the timetable and owns the response to any enquiry.