Module 2: What counts as R&D

After this module you will be able to say what makes work research and development for tax purposes, test a project against the two ingredients every definition shares, separate technical uncertainty from commercial and project risk, and recognise the activities the seven countries leave out.

What this module covers

• Why the tax definition reaches well beyond laboratories.
• The two ingredients every definition shares: a scientific or technological uncertainty, and systematic work to resolve it.
• Technical uncertainty versus commercial and project risk.
• The activities the seven countries exclude, and where a project starts and stops.

Wider than the laboratory

Most people picture R&D as laboratory science. The definitions say something different. They speak of a field of science or technology, and engineering, software, materials, food processing and manufacturing methods are all fields of technology. What matters is whether the work set out to achieve something the field did not yet know how to do, and went about it systematically. The guide on where qualifying work hides in manufacturing gives worked cases.

The two ingredients

A scientific or technological uncertainty

Several of the seven countries define the gap in almost the same words: uncertainty exists when knowledge of whether something is scientifically possible or technologically feasible, or how to achieve it in practice, is not readily available or readily deducible by a competent professional working in the field. The benchmark is the field, not the company: if an experienced practitioner could have worked out the answer, there was no uncertainty. And the test is applied at the start, so a project is judged by what was unknown when it began, not by whether it succeeded.

Most of the seven schemes also ask for an advance: new knowledge, or a new or improved product, process or service that the field did not have. The United States frames it differently: the work does not require novelty to the world and improving your own product counts, provided the uncertainty and the experimentation are present (see eligibility in the United States).

Systematic work to resolve it

The definitions describe a systematic progression: a hypothesis or design idea, an experiment or analysis to test it, observation, evaluation, and a conclusion that feeds the next step. Planned, recorded and evaluated trial and error meets this; tinkering that is never written down cannot be shown to. The outcome does not have to be success. A disproved hypothesis still generates knowledge, and a project that ends in failure can qualify if the records show what was tried.

Technical uncertainty versus commercial risk

The most common mistake in a first claim is to describe risk of the wrong kind: every new product carries uncertainty about customers, cost and timing, and none of that is scientific or technological.

The last row matters most for software. Guidance in several countries says the same thing: building a new product with established techniques and methods does not resolve a technological uncertainty, even where the result is commercially new.

What the countries exclude

Each country lists activities that are not R&D. Seven lists, one pattern: market research and sales promotion; quality control and routine testing; cosmetic or stylistic changes; research in the social sciences, arts or humanities; routine data collection; commercial production once the uncertainty is resolved; and, in some countries, software for the internal administration of the business. An exclusion removes an activity from the claim; it does not usually disqualify the project around it.

Where a project starts and stops

For claim purposes a project is the work done to resolve the uncertainty, not the product's whole life. The United Kingdom guidelines state it directly: R&D begins when work to resolve the scientific or technological uncertainty starts and ends when the uncertainty is resolved or work on it ceases. Production, maintenance, ordinary enhancements and support after that point are outside the claim even though the product continues. Drawing the boundary is a technical judgement; module 4 gives it to the technical lead and module 5 records the dates.

Illustrative example

Illustrative example. Harbour Robotics Ltd, a hypothetical company, is building a warehouse robot with three workstreams. The first is a gripper that must pick irregular soft items at line speed; no published design or supplier part does this, the engineers cannot predict which of four approaches will work, and they run recorded trials on each. That is a technological uncertainty judged against the field, addressed systematically: a candidate project. The second is the operator app, built with a standard framework and known methods, so it is not R&D, however important to the product. The third is a study of logistics customers to set the price: market research, excluded in all seven countries. Of the company's 100,000 of spending on the robot, only the gripper work goes forward to module 3.

Across the seven countries

Where to go next

• Qualifying versus non-qualifying activities.
• How to run a project identification workshop.
• R&D tax incentives for manufacturers.
• Eligibility: all guides on what counts as R&D.
• R&D Tax Fundamentals: the program overview. Module 3 attaches costs to the projects identified here.