1. Why Engineering Claims Are Typically Larger Than Software Claims
Software companies claim primarily on staff costs. Their qualifying expenditure is almost entirely wage-related: salaries, employer National Insurance contributions, and employer pension contributions for the developers involved in qualifying work. Materials are negligible.
Engineering companies have an additional layer of qualifying expenditure that software businesses do not. Physical materials used in prototype builds qualify as consumables. Components destroyed during testing qualify. Reagents and test substances qualify. The cost of failed test batches qualifies. In manufacturing environments, where prototype runs might use significant quantities of raw material, that consumable cost base can rival the staff cost element of the claim.
Capital equipment does not qualify under the R&D relief scheme itself, but the costs of running test rigs, the energy used in testing, and consumable supplies used in experimental processes all do. An engineering firm running extended physical testing programmes will typically see qualifying consumable costs in the range of 20 to 40 per cent of total qualifying spend, compared to effectively zero for a pure software business.
Beyond consumables, engineering projects frequently involve external specialists. Structural test houses, materials laboratories, university research groups, and specialist measurement consultants all appear in engineering R&D supply chains. Subcontractor costs at 65 per cent under the merged scheme add materially to the claim total. A 35-person engineering company might have five or six external organisations contributing qualifying work to its projects in a given year.
The combination of staff costs, consumables, and subcontractor contributions means that a typical qualifying engineering firm with 20 to 50 engineers will generate a larger raw qualifying expenditure figure than a software business of equivalent headcount. The 2022-23 HMRC data reflects this: the average manufacturing claim is approximately 60 per cent higher than the cross-sector SME average.
2. What HMRC Defines as R&D in Engineering
HMRC applies the same definition across all sectors. The definition originates in the DSIT guidelines and has been in place, largely unchanged, since 2004. A project qualifies if it seeks an advance in science or technology by resolving technical uncertainty. The advance must be in overall knowledge in the field, not just the company's own knowledge.
The Technical Uncertainty Test
HMRC's test asks: could a competent professional working in the relevant field of science or technology readily have resolved the technical problem from existing publicly available knowledge? If yes, the work is not qualifying R&D. If the answer is no, because the solution was not known, the problem required novel approaches, or standard methods were insufficient to meet the technical requirements, the work is a candidate for relief. The question is not whether the work was difficult, innovative for the company, or new to the client. It is whether the technical uncertainty was real and not resolvable from what was already known in the field.
In engineering, this test is applied to physical problems rather than digital ones. A structural engineer developing a connection detail for a novel composite material under dynamic load conditions may face genuine technical uncertainty if no validated design approach exists for that combination of material, geometry, and loading. A mechanical engineer designing a standard steel frame to a client specification using established codes is applying known solutions to a new application and is not doing qualifying R&D.
The key is the nature of the uncertainty. Engineering projects often involve both routine and novel elements. The qualifying work is the element where the team had to go beyond what existing knowledge and standard practice could offer. Everything else, however technically demanding, is not qualifying.
HMRC's guidance explicitly lists engineering as one of the sectors where qualifying R&D activity is common. The question for each project is not "is this engineering?" but "did this engineering project involve resolving genuine technical uncertainty?"
3. Qualifying Activity Types by Engineering Discipline
The following table sets out typical qualifying and non-qualifying activities across the main engineering disciplines. It is a guide, not a definitive list. Whether a specific activity qualifies always depends on whether genuine technical uncertainty was present.
| Discipline | Typical Qualifying Activity | Typical Non-Qualifying Activity |
|---|---|---|
| Mechanical | Novel mechanisms under extreme load or temperature; developing fatigue performance models for new material combinations; iterative prototype testing where failure mode was unknown | Standard component selection and sizing to known load cases; machine design using established supplier components and standard practice |
| Electrical / Electronics | Developing power management systems for new battery chemistries; designing high-frequency circuits where EMC behaviour could not be predicted; miniaturisation pushing beyond known assembly tolerances | PCB layout for known circuit topologies; standard sensor integration using manufacturer-specified interfaces; electrical installation to code |
| Civil / Structural | Foundation solutions for novel ground conditions outside published guidance; structural use of new or non-standard materials; behaviour modelling for structures under non-standard dynamic loads | Standard structural frame design to Eurocode; routine foundation design to established ground investigation data; compliance checking against known standards |
| Chemical / Process | Developing reaction conditions for new or modified chemical processes; scale-up from lab to pilot plant where behaviour cannot be predicted from small-scale data; novel catalyst development | Running established processes at production scale; optimising yield within known reaction parameters; quality control testing of known formulations |
| Software-embedded | Control algorithms for real-time systems where conventional approaches fail; signal processing methods for novel sensor types; firmware for hardware operating outside validated parameters | Standard PLC programming to a specification; configuring off-the-shelf SCADA systems; routine integration of industrial protocols |
| Aerospace | Structural solutions exceeding existing material and geometry validation envelopes; aerodynamic optimisation of novel geometries; thermal protection systems for new operating regimes | Applying existing certified designs to new configurations within validated parameters; standard systems integration to known interfaces |
| Defence | Hardening systems to threat environments where no design standard exists; novel materials for armour applications; signature reduction approaches for new platform geometries | Manufacturing to existing qualified designs; system integration to MIL-spec where solutions are fully defined |
4. The Prototype Question
Prototypes are a frequent and significant cost in engineering R&D claims, but HMRC's treatment of prototype costs requires careful handling. The rules hinge on two questions: is the cost capital or revenue in nature, and was the prototype built to resolve technical uncertainty or to produce a saleable product?
Where a prototype is built as part of a process of resolving technical uncertainty, the materials and labour costs involved are generally treated as revenue expenditure and can qualify as consumable or staff costs in an R&D claim. A prototype that fails, is dismantled, and contributes knowledge about what does not work is the clearest case. The failed prototype is evidence of the iterative process that defines qualifying R&D.
Where a prototype is the first production unit, built to demonstrate a known design rather than to test an unknown, it does not qualify. The distinction matters: a company that builds ten prototype iterations to refine a solution has a strong qualifying case for the materials and staff time in those iterations. A company that builds a prototype as a sales demonstration of a known design does not.
The capital versus revenue split is a separate consideration. If prototype tooling, jigs, or fixtures are created and retained as fixed assets, those capital costs do not qualify under R&D relief, though capital allowances apply. The consumable materials used in building the prototype, and the staff time involved in the build and testing, remain revenue expenditure and can qualify where the technical uncertainty test is met.
HMRC has confirmed that the subsequent sale or commercial use of a prototype does not automatically disqualify the development costs. Where the primary purpose of the build was to resolve technical uncertainty, the costs qualify even if the prototype was later sold or put to productive use. This is a point that engineering companies frequently miss: they assume that because a prototype ended up being used or sold, the costs cannot go into the claim. That assumption is wrong.
5. Consumables in Engineering Claims
Consumables are materials or substances directly used and consumed in the R&D process. In engineering, this category can encompass a wide range of expenditure. The key requirement is that the material must be directly used in the qualifying activity, not incidentally used or used in production work that happens to be near the R&D.
Materials that qualify include: raw materials used in prototype builds; test components that are stressed to failure; chemicals and reagents used in process development; coatings, adhesives, and finishing materials applied during prototype construction and testing; and specialist test media such as calibrated loads, fluids, or gases used in experimental rigs.
Consumable Cost Example: Process Engineering Prototype
A process engineering company develops a novel heat exchanger design for a corrosive chemical process. Over 18 months, the qualifying consumable costs include:
| Prototype tube materials (three iterations, two failed) | £18,400 |
| Specialist alloy test coupons for corrosion testing | £6,200 |
| Process fluids and reagents consumed in trials | £4,800 |
| Sealing and gasket materials (multiple configurations tested) | £3,100 |
| Non-destructive testing consumables | £2,400 |
| Total qualifying consumables | £34,900 |
Without a structured R&D assessment, this consumable spend would typically not appear in a claim. Added to the staff cost element of the same project, the consumables increase total qualifying expenditure for this project by approximately 35 per cent.
Consumable costs are sometimes overlooked in engineering claims because they are treated as project costs in the management accounts and are not obviously associated with R&D. A structured review of project cost records will typically surface consumable expenditure that the finance team had not considered as eligible.
6. Subcontractors: Common in Engineering
Engineering companies routinely use external organisations for specialist activities that cannot be performed in-house: environmental test chambers, fatigue testing facilities, specialist metrology laboratories, structural testing rigs, acoustic test facilities, and university research partnerships. These subcontractor costs qualify under the R&D relief scheme, subject to specific rules.
Under the merged scheme applicable from April 2024, qualifying subcontractor costs are included at 65 per cent of the amount paid. The discount reflects HMRC's view that the subcontractor itself may be claiming relief on the same work. The claimant company includes 65 per cent; the subcontractor may separately claim on its own qualifying costs.
For the costs to qualify, the subcontractor must be conducting work that would itself meet the definition of qualifying R&D, and the costs must be paid in the course of the claimant's R&D project. Routine testing against known acceptance criteria by a commercial laboratory does not qualify. Testing that forms part of resolving technical uncertainty, where the test house is effectively applying expert knowledge to a problem with an unknown outcome, does qualify.
University subcontractors and Knowledge Transfer Partnerships deserve particular attention. Engineering companies with university research collaborations often have significant qualifying subcontractor spend that is not being captured in claims. The academic relationship is sufficient; what matters is whether the work meets the technical uncertainty test.
Technical consultants engaged to help resolve specific engineering problems can also qualify as subcontractors. A specialist consultant brought in to work on a novel materials problem, a vibration challenge, or an aerodynamic issue where the company's own expertise was insufficient is a qualifying subcontractor cost provided the work meets the test.
7. What Engineering Companies Get Wrong
Common Errors in Engineering R&D Claims
- Claiming for production work. Staff time on production runs, manufacturing to a known process, and quality inspection of production output does not qualify. HMRC will ask hard questions about any claim where the qualifying percentage of time appears implausibly high for an organisation that also produces at volume.
- Including capital assets. Equipment, test rigs, and tooling purchased and retained as fixed assets cannot be claimed under R&D relief. The consumables used in operating that equipment can qualify. The purchase cost of the rig itself cannot.
- Missing subcontractor costs. Many engineering companies include only staff costs in their claims, leaving significant qualifying subcontractor expenditure unclaimed. Test houses, university partners, and specialist technical consultants often represent 20 to 30 per cent of total qualifying spend for engineering businesses.
- Underestimating the qualifying percentage. Engineers frequently underestimate how much of their time was spent on qualifying activities because the R&D is interwoven with project delivery. A structured technical interview process typically surfaces a higher qualifying proportion than an initial self-estimate by project managers.
- Not claiming for failed projects. A project that did not achieve its technical objective may be the most clearly qualifying project in the portfolio. HMRC's definition does not require success. Failed attempts that resolved uncertainty by demonstrating what does not work are qualifying R&D.
8. Worked Example: 35-Person Mechanical Engineering Firm
Worked Example: Qualifying Expenditure Calculation
Company profile: 35-person mechanical engineering consultancy. Annual turnover £4.2m. Work includes novel industrial equipment design, custom automation systems, and bespoke structural testing equipment for clients in advanced manufacturing.
Qualifying projects identified (one accounting year): Three qualifying projects involving novel actuation mechanisms, a custom environmental testing rig with no commercial precedent, and a high-speed assembly system where conventional approaches failed initial trials.
| Cost category | Total cost | Qualifying % | Qualifying amount |
|---|---|---|---|
| Staff costs (salaries, NI, pension) | £1,840,000 | 22% | £404,800 |
| Consumables (materials, test components) | £310,000 | 35% | £108,500 |
| Subcontractors (test house, university partner) | £185,000 | 60% qualifying, then 65% rule | £72,150 |
| Software licences (FEA, CFD tools) | £42,000 | 70% | £29,400 |
| Total qualifying expenditure | £614,850 |
Claim value under merged scheme (20% above-the-line credit):
| R&D credit at 20% | £122,970 |
| Corporation tax on credit (25%) | (£30,743) |
| Net benefit (profitable company) | £92,228 |
This firm had not previously claimed R&D tax credits. The qualifying percentage of staff time (22%) reflects engineers spending approximately one day per week on qualifying technical development activity across the three projects, alongside their standard client delivery work.
9. HMRC Enquiry Risk in Engineering
Engineering claims carry above-average enquiry risk for a straightforward reason: claim values are larger. HMRC's compliance teams apply risk-based selection criteria, and higher-value claims receive proportionally more scrutiny. A manufacturing company claiming £150,000 will face more examination than a software startup claiming £40,000.
The documentation requirements are not different for engineering than for other sectors, but the standards need to be met with particular care. HMRC will want to understand: which specific projects qualified; what the technical uncertainty was at the outset; what approaches were tried; how the company knows whether the uncertainty was resolved; and how the qualifying costs were identified and separated from non-qualifying costs.
In engineering, the technical uncertainty narrative needs to be written by someone who understands engineering. A claim that says "we designed a new machine" will attract an enquiry. A claim that explains the specific material behaviour, loading condition, or process parameter that was unknown, what approaches failed, and what was ultimately established is a claim that stands up to scrutiny.
Documentation Standards for Engineering Claims
- Project technical scope. A clear statement of the technical objective and the specific uncertainty that existed at the outset. This does not need to be a formal R&D document; engineering project briefs, technical proposals, or specification documents that describe the technical challenge are acceptable.
- Evidence of iteration. Design review records, test reports, meeting notes, or internal technical communications that show the iterative process. Failed attempts are as important as successes.
- Cost records by project. Timesheets, material purchase records, and subcontractor invoices that can be allocated to specific qualifying projects.
- Competent professional assessment. The technical narrative in the claim should reflect an assessment of what a competent professional in the field would have been able to resolve from existing knowledge. A specialist R&D adviser experienced in engineering claims will conduct technical interviews with the engineering team to build this narrative correctly.
Engineering companies that use a specialist adviser experienced in their sector consistently achieve higher claim values and lower enquiry rates than those that use a generalist or attempt to self-prepare. The investment in competent claim preparation is proportionate given the values at stake.
Frequently Asked Questions
It can, but the test is whether the design work involved resolving genuine technical uncertainty. Routine design to a client specification, applying standard techniques and known solutions, does not qualify. Design work that required developing new approaches, solving problems no standard method could address, or iterating through failed attempts to reach a technical solution may qualify.
Yes, in some cases. Where the consultancy bears the financial risk of the R&D activity and the technical uncertainty arises from the nature of the problem rather than the client specification, qualifying work within client projects can support a claim. The contractual structure matters significantly and a specialist adviser will assess the arrangements.
Licences for CAD, FEA, CFD, and other simulation tools used directly in qualifying R&D work can qualify as software expenditure. General-purpose design software used across all client work would need to be apportioned to reflect only the qualifying use.
HMRC expects evidence of technical uncertainty at the start of the project, the methods tried, and the outcome. In engineering this typically means project documentation, design iteration records, test reports, and internal communications about technical challenges. No specific format is prescribed but the records must demonstrate a systematic process of resolving genuine uncertainty.
Yes, provided the costs were incurred in resolving technical uncertainty rather than simply producing a saleable product. Where a prototype was built to test an approach and the primary purpose was resolving a technical question, the costs can qualify even if the prototype was subsequently sold. The revenue from the sale does not offset the qualifying expenditure.
Yes, where the test house is carrying out qualifying R&D on your behalf. Under the merged scheme, subcontractor costs qualify at 65 per cent of the amount paid. The test house must be conducting work that would itself qualify as R&D, not simply performing routine compliance testing against known acceptance criteria.