What the First-tier Tribunal’s decision in Tanglewood Care Services tells us about evidencing an R&D claim
On 6 August 2026, the First-tier Tribunal dismissed an appeal by Tanglewood Care Services Limited against HMRC’s decision to remove an R&D tax relief claim worth £880,286. The company operates seven residential nursing homes. The claim covered its response to Covid-19 for the year to 31 January 2021 — including testing regimes, PPE protocols, visitor restrictions, resident cohorting, enhanced cleaning, staff segregation, and track-and-trace.
At the outset, it is worth saying what this case is not. The Tribunal found all three of the company’s witnesses to be “honest and conscientious.” No one suggested the work had not taken place or that it had not mattered. Tanglewood was protecting very vulnerable people during the worst public health emergency in a century, and it did so with real ingenuity.
The claim still failed. Understanding why is useful for anyone claiming R&D relief, in any sector.
Uncertainty in the world is not uncertainty in your project
The company’s strongest argument was also the most intuitive: in early 2020, nobody knew how to run a care home safely during a novel pandemic. Government guidance changed constantly, and the science was unsettled. If that isn’t uncertainty, what is?
The Tribunal agreed with the premise but rejected the conclusion. It accepted that the period “was characterised by considerable uncertainty” and that “Covid-19 was a novel disease, scientific understanding evolved rapidly and public-health guidance changed over time” (para 103). However, it drew a line that every claimant should understand:
“Uncertainty within the wider scientific and public health community does not, of itself, establish that the Appellant’s activities were directed towards resolving scientific or technological uncertainty” (para 105)
And more pointedly:
“The Appellant was seeking to determine the best way to respond to evolving scientific knowledge, public health advice and practical experience within its homes. That is not the same as resolving a scientific or technological uncertainty within the meaning of the Guidelines” (para 108)
This is the distinction between operating in uncertain conditions and setting out to resolve uncertainty. It covers more claims than people expect. A volatile market, a new regulation, an unproven supplier, a shifting standard — these make a project difficult. They do not make it R&D. The question is not “was the answer unclear?” but “did the company not know how to do something and set out to find out?”
Combining known things can qualify — but the combination must be going somewhere
Tanglewood’s second argument was that the individual measures should not be examined in isolation. Each measure was known; the difficulty lay in how they interacted within a care home. This is the “system uncertainty” point at paragraphs 29–30 of the DSIT Guidelines, and it is a legitimate and frequently used route to a claim.
Encouragingly, the Tribunal accepted it as a matter of principle:
“The Guidelines are broad enough to encompass system uncertainty arising from the interaction of multiple measures within a system. A claimant is not confined to uncertainties concerning the behaviour of individual scientific or technological components” (para 104)
It also confirmed — helpfully, for software and process claims — that an advance need not be an advance in underlying knowledge. Paragraph 13 of the Guidelines “expressly contemplates uncertainty arising from turning something already known to be scientifically feasible into a cost-effective, reliable and reproducible process, product or service” (para 94).
So the door is open. What Tanglewood could not do was walk through it. The Tribunal found that the objective was “to determine how best to deploy, balance and manage infection-control measures within the Appellant’s own care homes” and that the activities “were not directed towards achieving an advance in overall knowledge or capability beyond the Appellant’s own operations” (para 99).
A combination claim needs to identify the capability the integration was intended to advance — one that matters beyond your own four walls. Solving your own problem exceptionally well is not, on its own, sufficient.
The competent professional must be someone who can speak to the field
The most transferable lesson concerns evidence. None of Tanglewood’s witnesses had qualifications in epidemiology, virology or infection science. The company argued that the relevant competent professionals were experienced care-home operators and nursing professionals, rather than specialist virologists.
The Tribunal’s response is careful, and worth reading closely, because it is not the crude “you needed a virologist” that it is sometimes reported as:
“The absence of evidence from a person able to speak authoritatively in the relevant field is significant, although not because we consider the relevant field must necessarily have been virology or epidemiology. Rather, it leaves the Tribunal without sufficient evidence to determine whether the alleged uncertainties were scientific or technological, whether they were readily deducible by a competent professional, or whether the activities sought an advance in overall knowledge or capability beyond the Appellant’s own operations” (para 113)
The point is not about credentials. The competent professional establishes what was already known (the baseline) and therefore determines whether anything was genuinely uncertain. Without that person, the Tribunal has no way to test the claim. The competent professional is not a name to be entered in a box on the Additional Information Form. They are a witness.
On documentation, the Tribunal was more generous than you might expect
Tanglewood filed its corporation tax return in February 2022, with no R&D claim. The amended return, carrying the £880,286 claim, arrived in September 2022, some seven months later. There was no formal project plan; decisions were made through ongoing management discussions and adapted as events unfolded.
HMRC placed much emphasis on this, submitting that the R&D report was prepared after the accounting period and “did not identify the relevant baseline knowledge, the uncertainties to be resolved, any methodology for resolving them, or criteria by which success could be assessed” (para 69).
The Tribunal did not adopt that argument. On the contrary:
“Although no formal project plan existed and many decisions were taken incrementally as circumstances evolved, we do not consider that paragraph 19 requires a formally documented plan.” (para 87)
That is worth holding on to, as it cuts against widespread anxiety. Real R&D is often messy and rarely documented to a tribunal’s standard while it is happening, and this decision confirms that a missing project plan does not, by itself, sink a claim. Contemporaneous records are evidence, not a statutory condition.
But there is a sting. Good records can also fill an evidential gap — and Tanglewood had neither. When a claim cannot produce a competent professional to establish what was already known, contemporaneous documents showing what the company did not know and what it did to find out become the next best proof. Records are not required. They are, however, extremely useful when the rest of your evidence is thin.
What this decision does not say
First-tier Tribunal decisions are not binding precedent, and this one turns heavily on its own facts and a thin evidential record. It does not put the care sector off-limits. It does not close down system or combination claims — if anything, paragraphs 94 and 104 are mildly helpful to them. And it does not require a PhD in the witness box.
What it does is describe, unusually clearly, what a tribunal needs to see: a baseline that a qualified person can speak to, an uncertainty the company set out to resolve rather than merely live with, and an advance that means something beyond the claimant’s own operations.
If you are preparing a claim now
Three questions are worth asking of any narrative before it goes to HMRC:
- Does the uncertainty section describe a gap in the company’s capability or conditions in the external environment? If it reads as a description of a difficult environment, it needs rewriting.
- If the claim rests on integrating known components, what capability was the integration intended to advance? Name it and explain why it extends beyond this company.
- Could your competent professional provide evidence on the state of knowledge in the field? Not “are they senior” — could they be cross-examined on the baseline and hold up?
If the answers are uncomfortable, it is far better to discover that now than in a closure notice three years from now.
Vantage R&D Consulting prepares R&D tax relief claims for UK SMEs. If you would like a second opinion on a claim in preparation, or on one currently under enquiry, we are happy to take a look.
Tanglewood Care Services Limited v HMRC [2026] UKFTT 1137 (TC), 6 August 2026. The full decision is available on the National Archives Find Case Law service: https://caselaw.nationalarchives.gov.uk/ukftt/tc/2026/1137