How to Identify Your Competent Professional

An R&D claim rests on expert judgement, and often on more than one expert. The baseline, the advance and the scientific or technological uncertainties you describe on the Additional Information Form all come from a competent professional’s assessment of the work. If the people giving that assessment do not meet HMRC’s test, or do not between them cover the fields the project actually touched, there is nothing underneath the narrative.

The Additional Information Form does not ask you to name them. That has led some companies to treat the competent professional as a formality. HMRC does not. In Part 3 of its Guidelines for Compliance 3 (GfC3), HMRC states that it has reduced the value of some claims to zero where a competent professional had not identified that the company was seeking an advance in science or technology.

Choosing the right people is a decision, not an administrative step. This blog sets out what HMRC’s test requires, why one project can need more than one competent professional, the two ways companies most often get the choice wrong, and five questions that will tell you quickly whether the person in front of you can carry their part of the claim.

Who this applies to

This is for directors deciding who should stand behind the technical narrative on an R&D claim, and for finance teams who have been told the managing director will handle that part. It matters most in engineering, manufacturing and software businesses where technical knowledge sits with specific individuals rather than with the board, and most of all where a project crosses disciplines.

It applies whether you are preparing a first claim or reviewing one you have made for years. If you cannot name the competent professional behind each uncertainty in your last claim, and say why that person rather than another, the question is live for you now.

What HMRC’s test actually requires

The statutory definition of R&D turns on what a competent professional working in the field would, or would not, find readily deducible. That test runs through the DSIT Guidelines on the meaning of R&D for tax purposes, issued on 7 March 2023 and replacing the 2004 guidelines, and through HMRC’s own manual at CIRD81300. The person is not a supporting detail. The person is the measuring instrument.

GfC3 describes a competent professional as someone suitably qualified or experienced in the field, and expects them to hold all three of the following:

• Knowledge of the relevant scientific or technological principles involved.
• Awareness of the current state of knowledge in the field as a whole, not only within your business.
• Accumulated experience, with a successful track record in that field.

HMRC then closes off the most common shortcut in a single sentence. Having worked in a field, or having an intelligent interest in it, does not on its own make a person a competent professional. Length of service is not the test. Neither is seniority.

The specialism trap

The most common failure is not a lack of expertise. It is expertise in the wrong field.

GfC3 gives an example of a director with extensive experience and recognition in the wholesale trade, who is put forward as the competent professional for a project developing stock management IT systems. HMRC’s view is that his competency lies in the wholesale trade, not in the specialist area of stock management IT systems. He is an expert. He is not the expert this project needs.

The same guidance recognises the opposite situation, where an in-house team is too inexperienced to identify technological uncertainty without bringing in external expertise. Both examples point the same way. The field of the project decides who qualifies, and the field is narrower than the industry.

Why one project often needs more than one

The specialism trap has a second form, and it appears inside a single project rather than across a claim.

Paragraph 20 of the DSIT Guidelines sets the baseline as the overall knowledge or capability in a field of science or technology, meaning what is publicly available or readily deducible from it by a competent professional working in the field. The test is framed one field at a time. Paragraph 19 then defines a project as all the activities that collectively serve to resolve the uncertainty, which it says may include a number of different sub-projects.

Put those two together and the consequence is practical. A project whose sub-projects sit in different disciplines has more than one baseline, and each baseline has to be established by someone who genuinely holds current knowledge of that field as a whole. A materials problem and a control software problem inside the same development programme are two fields, not one. Asking a single person to speak to both is the wholesale trade example again, only harder to spot because the person is at least in the right industry.

HMRC’s own materials use the plural more often than companies notice. Its guidance on the additional information you must submit refers to supplying details of the competent professionals in a supporting report, and the GfC3 examples describe company engineers, plural, as competent professionals. Nothing in the guidance limits a project to one.

The practical order is therefore to identify the fields the project touched before identifying the people. Working the other way round, starting with whoever is available and asking what they can cover, is how a claim ends up with an uncertainty nobody is qualified to have described.

Two failure modes, and why both are expensive

Assuming you have found someone in the right discipline, they will usually sit at one of two poles. Both distort a claim, in opposite directions.

The engineer who says nothing qualifies

Conservative, precise, and inclined to treat anything they solved as obvious in hindsight. Ask whether the project involved uncertainty and the answer is that it was just engineering. This is the more credible witness and the less reliable narrator, because hindsight is exactly what the test excludes. The question is not whether the answer looks obvious now. It is whether it was readily deducible at the time by a competent professional working in the field.

The engineer who says everything is R&D

Enthusiastic, commercially proud, and inclined to describe difficulty as uncertainty. Ask what was already known in the field and the answer turns back to how hard the project was, how tight the deadline, how demanding the customer. Difficulty is not the test either. A project can be gruelling, expensive and late without containing a single technological uncertainty.

The first costs you a claim you were entitled to make. The second costs you a claim you were not, plus the enquiry that follows. Neither person is being difficult. They are answering a question that is easy to ask badly. Where a project spans disciplines you will often find one of each, which is a further reason not to let a single voice set the boundaries for the whole project.

Five questions that tell you which one you have

These are the questions we use. They are deliberately not variations of “was this R&D”, because that question invites the answer the person is already inclined to give.

1. What would a competent professional elsewhere in your field have known about this at the time? A usable answer describes the state of the field. An unusable answer describes your company.
2. What did you look at before you started, and what did it tell you? Standards, published methods, supplier documentation, academic literature, patents. A competent professional can say what they consulted and why it did not answer the question.
3. At what point did you stop knowing the answer? This locates the uncertainty on a timeline. If the person cannot identify that point, either it did not exist or the project boundary is drawn in the wrong place.
4. What did you try, in what order, and why did each approach change? Systematic iteration is the signature of resolving uncertainty. A single approach that worked first time usually is not.
5. What would have made this not R&D? The most revealing question of the five. Someone who understands the test can describe the version of this project that would have failed it. Someone who cannot has not been applying a test at all.

Ask them once per field rather than once per project. If someone answers the first question crisply on the mechanical side and vaguely on the software side, that is not evidence the software work fails the test. It is evidence you are talking to the wrong person about that half of it, and that a second competent professional is needed rather than a broader answer from the first.

None of these questions requires a tax specialist to ask. They do require someone in the room who can tell whether the answers are true, which is a different skill from knowing the legislation.

Why the director’s enthusiasm is not evidence

Directors are frequently the ones who raise R&D relief, who describe the project in the first meeting, and who are most confident it qualifies. That confidence is genuine and it is not evidence. A director who approved the budget and chaired the reviews knows what the project cost, what it was for, and what it meant commercially. Those are the wrong three things.

This is also why we ask for the project lead by name rather than assuming the director is the right person. In smaller businesses the two are sometimes the same individual, and a hands-on engineering director with current knowledge of the field can be a perfectly good competent professional. GfC3 gives an example of exactly that. The point is that the role does not confer the status. The knowledge does, and it does so one field at a time.

Two candidates for the same claim

	Candidate A	Candidate B
Role	Managing director, 30 years in the sector	Senior process engineer, 11 years in this discipline
Knowledge of the field	Broad commercial and operational	Current on published methods and standards
Involvement in the project	Approved the budget, chaired reviews	Set the approach, ran the trials
Can state what was already known	No	Yes, with sources
Suitable as competent professional	No	Yes, for the process engineering work

Candidate A is not disqualified from the process. He is the right person to describe why the project mattered to the business, to confirm who worked on it, and to approve the claim. He is not the person whose knowledge the statutory test is measured against.

Note the qualification in the last row. Candidate B is the competent professional for the process engineering, and only for that. If the same project also involved embedded control software, the claim needs a third name, and the absence of one is not a gap in the paperwork. It is a gap in the evidence for that uncertainty.

Worked example: what the wrong choice costs

A manufacturing company prepares a claim covering two projects, with qualifying expenditure of £480,000 in total. Project 1 accounts for £180,000 and Project 2 for £300,000. Within Project 2, £180,000 relates to process engineering and £120,000 to the embedded control software developed alongside it. Assume an accounting period beginning after 1 April 2024, so the merged scheme applies, the company is profit-making and pays Corporation Tax at the 25% main rate.

Under the merged scheme the expenditure credit is 20% of qualifying expenditure and is itself taxable, so £480,000 produces a gross credit of £96,000 and a net benefit of around £72,000 on these assumptions.

Two things can go wrong here, and they cost different amounts.

• Nobody qualified speaks to Project 2 at all. The managing director describes a demanding customer specification, a compressed timeline and a successful launch. The whole £300,000 is exposed, which is £60,000 of gross credit and roughly £45,000 of net benefit.
• One competent professional covers only part of it. The process engineer gives a strong account of the materials and tooling work but has no standing to describe the state of knowledge in control software, and no one else is asked. The £180,000 he covers holds. The £120,000 he does not is exposed, which is £24,000 of gross credit and roughly £18,000 of net benefit.

The second scenario is the more common and the harder to spot, because the claim has a named, well-qualified competent professional attached to it and reads convincingly. Bringing in a second, the software lead, closes that £120,000 gap at the cost of one conversation.

The figures are illustrative and the rates are those applying on these assumptions, but the shape holds generally. The exposure is not spread evenly across the claim. It concentrates where the wrong person did the talking, or where nobody was asked to.

Where a second competent professional strengthens a claim

Coverage across disciplines is the case where a second name is necessary. There are three further cases where it is not required but, in our experience, materially strengthens the position. These are our view rather than an HMRC requirement, and we would distinguish them clearly from the guidance above.

• Corroborating the baseline. The baseline is the weakest link in most technical narratives, because it is an assertion about what an entire field already knew. Two practitioners independently describing the same state of knowledge, with the same sources, is considerably harder to dislodge than one person asserting it, particularly when a caseworker produces a published source and argues the advance was readily deducible.
• Continuity. Claims are often examined two or three years after the work. A single competent professional is a single point of failure if that person leaves, retires or simply stops remembering the detail.
• Distance from the cost line. Where the only competent professional is also the individual whose time makes up most of the apportionment, their account of the project supports the expenditure they themselves generated. That is not improper and it is very common, but a second view from someone outside the cost schedule carries more weight when the claim is tested.

There is a limit to this, and it matters. More names do not produce more relief, and a list of people without evidence of their competency is weaker than one well-evidenced professional. Each person must genuinely hold current knowledge of the field they are speaking to, and must have engaged with the project rather than reviewed a summary of it. A second opinion from someone who does not meet the test adds nothing to the claim and gives HMRC another thread to pull.

Evidencing the people, not just the project

Having identified the right people, record why each of them is the right person. GfC3 sets out what HMRC accepts as evidence of competency:

• High-level qualifications in the field, together with continuing professional development.
• Significant years spent working at a high level in the field.
• A record of scientific or technical publication.
• Industry awards, or public recognition for contribution to the field.

Where more than one person is involved, record which field and which uncertainty each of them covered. A claim naming three competent professionals without saying who spoke to what is harder to defend than one naming a single professional with a clearly bounded scope.

HMRC also expects the competent professional’s opinion to do more than assert that a project qualifies. It should set out their own depth of knowledge and experience, the current state of knowledge in the field, the advance being sought, why it is an advance, and whether it relates to knowledge, capability or both. Our nine-part framework for R&D project records sets out where those opinions sit in the wider project record, and our record-keeping checklist for engineering projects covers the surrounding evidence trail.

What if the person has left

This is common and it is not fatal. The competent professional does not have to be a current employee, and HMRC’s expectations of claimants contemplate drawing on external expertise where the necessary knowledge is not held in house. What matters is that the person giving the opinion has engaged with this project, not a summary of it.

Where the original engineer has moved on, the practical options are to approach them for an account of the work, or to have a suitably qualified colleague or external specialist review the contemporaneous project evidence and give their own opinion on it. The second is weaker than the first and both are considerably weaker than having asked at the time. This is the strongest practical argument for identifying more than one competent professional at the outset, and for capturing both accounts while the work is happening rather than at the point of claiming.

How Lexmore approaches this

We identify the fields a project touched first, then the competent professional for each, as part of the eligibility work, before any costs are gathered and before the letter of engagement is signed. That means asking for the project lead by name rather than accepting whoever is available, putting the five questions above to each person in their own field, and flagging where an uncertainty has no one qualified attached to it. Where the answers do not establish an advance and an uncertainty, we say so, and the conversation ends there rather than continuing into a costing exercise.

Some of the work we review qualifies. A good deal of it does not, including work that is genuinely difficult and commercially valuable. Establishing which is which is the point of the exercise, and it is easier to do before a claim is filed than during a compliance enquiry.

Lexmore’s View

The competent professional is the load-bearing element of an R&D claim. Everything else, the narrative, the boundaries, the apportionment, is built on an assessment of what the field already knew. HMRC has been willing to reduce claims to nil where that assessment was never made, which tells you how it weighs the point.

The question worth asking is not simply who your competent professional is. It is whether the people you have named cover every field the work touched, and whether each of them can evidence the baseline they are asserting. Identifying them is the first substantive decision in the process, not a formality to complete once the claim is drafted. The right people. The right questions. A claim that holds.

References

• GfC3 Part 3: Importance of a competent professional
• GfC3 Part 2: Expectations of claimants
• DSIT: Meaning of research and development for tax purposes (7 March 2023)
• CIRD81300: The definition of R&D for tax purposes
• HMRC: Additional information you must submit before you claim R&D tax relief