The short answer is yes. A company can claim R&D tax credits for a failed project that never generated revenue. HMRC does not require commercial success. The legislation focuses on the attempt to overcome technical uncertainty, not on whether the project made money.

This is one of the most common misconceptions we see as ICAEW qualified accountants at Holloway Davies. Directors of small and growing limited companies often abandon legitimate R&D claims because the project did not work. They assume the tax relief only applies to successful innovations. That is wrong, and it costs them real money.

This article explains exactly what qualifies for R&D tax credits when a project fails, how HMRC assesses these claims, and what evidence you need to hold onto.

What HMRC Actually Looks For in an R&D Claim

HMRC’s definition of R&D for tax purposes comes from the Guidelines on the Meaning of Research and Development for Tax Purposes (BIS 2004, updated 2010). It does not mention commercial success anywhere in the criteria.

The test has two parts:

  • Technical uncertainty. Could your project team know at the start whether the outcome was achievable, or how to achieve it, based on readily available information?
  • Attempt to resolve. Did you undertake a systematic, investigative, and iterative process to try to resolve that uncertainty?

If the answer to both is yes, the work qualifies as R&D. Whether the project generated revenue, found a market, or was ultimately abandoned is irrelevant to the definition.

Think of it this way. A pharmaceutical company spends eight years developing a drug that fails in phase 3 trials. They write off millions. HMRC allows the R&D claim. The same logic applies to a small software consultancy in Manchester that spends £40,000 trying to build a novel machine learning algorithm that never worked. The work still qualifies.

Real Examples of Failed Projects That Qualified

We have handled claims for failed projects across multiple sectors. Here are anonymised examples that show the pattern.

A Bristol Engineering Firm

A 12-person engineering company in Bristol spent £92,800 trying to develop a new composite material for marine components. The material failed structural testing after 14 months of work. No product ever went to market. HMRC accepted the full R&D claim because the team faced genuine technical uncertainty about the material's behaviour under stress, and they conducted systematic testing to resolve it.

A Shoreditch Software Startup

A five-person software company in Shoreditch spent £63,400 building a real-time data processing platform that could not achieve the required latency. The project was abandoned after nine months. The claim was accepted because the technical challenge (sub-millisecond processing across distributed nodes) was not known to be solvable at the start, and the team tried multiple architectures before concluding it was not feasible with existing hardware.

A Birmingham Food Science Company

A husband-and-wife limited company in Birmingham spent £27,500 developing a plant-based protein stabilisation method that never worked at commercial scale. HMRC accepted the claim. The technical uncertainty was whether the stabilisation could survive pasteurisation without degrading flavour. It could not. The claim stood.

In every case, the key factor was that the work attempted to resolve a genuine technical uncertainty. Commercial failure did not matter.

The Technical Uncertainty Test: How to Apply It

The technical uncertainty test is the core of every R&D claim, successful or not. You need to show that competent professionals in your field could not have known the solution at the outset.

This is not the same as commercial uncertainty. "We did not know if customers would buy it" is not R&D. "We did not know whether the material could withstand 200 degrees Celsius without degrading" is R&D.

For a failed project, the technical uncertainty test often becomes easier to demonstrate. The fact that the project failed is itself evidence that the outcome was not certain. But you still need to show that the uncertainty was technical, not commercial or operational.

Ask yourself these questions:

  • Was the problem genuinely novel, or were you applying known methods to a standard problem?
  • Could a competent professional in your field have told you the answer upfront?
  • Did you document the uncertainty before starting the work?

If the answer to the first two is "yes" and you have written evidence of the third, you have a strong claim.

What Evidence Do You Need for a Failed Project Claim?

HMRC will scrutinise claims for failed projects more closely than claims for successful ones. That is simply because failed projects leave less obvious commercial evidence. You need to compensate with strong technical documentation.

Keep the following records:

  • Project initiation documents. Emails, meeting notes, or project briefs that describe the technical problem at the start. These prove the uncertainty existed before the work began.
  • Technical logs. Lab notebooks, development logs, test results, error reports. Show what you tried and what happened.
  • Decision records. Notes explaining why each approach was abandoned. "We tried X, it failed because Y, so we moved to Z."
  • Time records. Who worked on the project, for how many hours, and what they did. HMRC will ask for this.
  • Cost records. Staff salaries, subcontractor costs, materials, software licenses. The usual R&D qualifying costs.

For a failed project, the narrative is critical. You need to tell the story of the technical journey, not just the outcome. A well-written technical report that explains what you tried, why it failed, and what you learned is the strongest evidence you can provide.

How the Claim Amount Is Calculated

The calculation works exactly the same way as for a successful project. The qualifying costs are the same categories: staff costs, subcontractor costs (65% of the payment), consumables, software, and externally provided workers.

For accounting periods starting on or after 1 April 2024, the merged R&D scheme applies to most companies. The calculation depends on whether your company is profit-making or loss-making.

Profit-making companies. You deduct the qualifying R&D expenditure from your trading profits before calculating corporation tax. For a company paying 25% corporation tax, every £10,000 of qualifying R&D spend saves £2,500 in tax. At the 19% small profits rate, it saves £1,900.

Loss-making companies. If the R&D spend creates or increases a trading loss, you can surrender that loss for a cash payment. The payable credit rate depends on your company's R&D intensity. For non-intensive companies, the rate is 15% of the surrenderable loss. For R&D intensive companies (where qualifying R&D spend is 30% or more of total costs), the enhanced R&D intensive scheme (ERIS) applies, with a higher payable credit rate of 27% for loss-making companies.

A failed project that never generated revenue is likely to leave your company loss-making on that activity. That makes the payable credit route the most relevant option.

Let us use a worked example. A four-person software company in Leeds Dock spends £74,500 on a failed R&D project. Staff costs are £52,000, subcontractor costs are £14,000 (65% of £21,538), and consumables are £8,500. Total qualifying spend: £74,500. The company has no other revenue. Under the merged scheme, the company surrenders the loss and receives 15% as a cash credit: £11,175. That is cash in the bank from a project that made zero revenue.

Common Mistakes When Claiming for Failed Projects

We see the same errors repeatedly. Avoid these.

Not claiming at all. The biggest mistake. Directors assume failure disqualifies the work. It does not.

Poor documentation. HMRC will ask for evidence. If you cannot show the technical uncertainty existed at the start, the claim may be rejected. Start documenting from day one, even if you are not sure the project qualifies.

Confusing commercial and technical uncertainty. "We did not know if the market would accept it" is not R&D. "We did not know if the material would hold its shape at 150 degrees" is R&D. Be precise.

Claiming for routine problem solving. If a competent professional in your field could have solved the problem without research, it is not R&D. Failed routine work is still not R&D.

Ignoring the R&D tax credits notification requirement. For accounting periods starting on or after 1 April 2024, you must notify HMRC within 6 months of the period end if you intend to claim. Miss this deadline and the claim is invalid. The notification is made through the CT600 (corporation tax return) and the Additional Information Form (AIF).

What Happens If HMRC Opens an Enquiry?

HMRC may open an enquiry into any R&D claim, but failed projects attract more attention. That does not mean you should not claim. It means you should prepare properly.

If HMRC opens an enquiry, they will ask for the technical narrative and supporting evidence. Our ICAEW qualified team handles these enquiries regularly. The key is to have the documentation ready before you submit the claim, not after HMRC asks for it.

A well-documented failed project claim is defensible. A poorly documented one is not. The difference is usually a matter of a few hours spent writing up the technical narrative at the time of the work, rather than trying to reconstruct it three years later.

If HMRC rejects the claim, you can appeal through the usual tribunal process. But the vast majority of properly documented claims are accepted, even for failed projects.

When Should You Speak to an Accountant About a Failed Project Claim?

If your company spent money on a technical project that did not work, and you are unsure whether the work qualifies as R&D, speak to an accountant before you file your CT600. The notification deadline is strict, and missing it costs you the claim entirely.

We offer a free initial review of potential R&D claims. You send us the project details and costs. We tell you whether the work qualifies and what the likely claim value is. No obligation. Contact us if you want to run a project past us.

The key takeaway is this. HMRC rewards the attempt to resolve technical uncertainty, not commercial success. A failed project that never generated revenue can still be a valuable R&D claim. Do not leave that money on the table.