Qualifying versus non-qualifying activities: where the line really sits

The four schemes we cover use different words for the same idea. The United States asks for a process of experimentation to eliminate uncertainty. The United Kingdom asks for an advance in science or technology through the resolution of scientific or technological uncertainty. Australia asks for experimental activities whose outcome cannot be known in advance. Canada asks for systematic investigation to resolve scientific or technological uncertainty.

Strip away the vocabulary and each scheme is asking two questions: was there a genuine technical unknown, and did you work on it systematically? The work that answers "yes" to both is inside the line. Everything else, however necessary to the business, is outside it.

Inside the line

• Designing, building and testing a prototype to find out whether an approach works.
• Developing a new algorithm, material, process or architecture where the existing literature and the team's expertise did not already provide the answer.
• Systematically trialling alternatives, recording results and revising the approach.
• Engineering, design and testing work carried out to resolve the unknown, including failed attempts.
• Directly supporting work that would not have happened without the R&D, such as building a test rig, within the limits each scheme sets for work that supports the R&D.

Outside the line

• Routine engineering, maintenance and bug fixing using known methods.
• Configuring, installing or integrating purchased software or equipment following vendor documentation.
• Quality control, routine testing and commercial production once the technical unknown is resolved.
• Market research, sales, user research and commercial feasibility.
• Cosmetic, aesthetic and stylistic changes.
• Management, administration and training that is not specific to the R&D.
• Work in the social sciences, arts and humanities, which all four schemes exclude.

Sector examples

Software

Building a customer portal on a standard framework is outside the line. Developing a new approach to synchronising data across unreliable connections, where the standard techniques could not meet the latency and consistency requirements, is inside it. The portal features built on top of the new sync layer are outside again.

Manufacturing

Buying and commissioning a new machine is outside. Developing a process to run a material through it at a speed or tolerance the manufacturer said was not possible, through trials and adjustment, is inside. Producing saleable product on the finished process is outside.

Engineering and construction

Designing a building to code is outside. Developing a structural or environmental solution for a site condition that existing methods could not address, with modelling and testing, is inside. The rest of the build is outside.

Food and consumer products

Changing a flavour is outside. Reformulating to remove an ingredient while maintaining shelf life and texture, where no known substitute achieved it and trials were needed, is inside. Scaling the finished recipe for production is usually outside, unless scale-up itself raised a new technical unknown.

Where R&D begins and ends

R&D starts when work to resolve the unknown begins and ends when it is resolved or abandoned. The uncertainty is the boundary in time as well as in scope. A good test: if the technical unknown vanished tomorrow, which work would stop? That work is your R&D.

Grey areas and how to decide

Some work sits close to the line: extensive customisation of a platform, systems integration at unusual scale, or applying a known technique in a new field. The decision rests on whether a competent professional in the field, with access to the public state of knowledge, could have known how to do it without experimentation. Write the answer down, with the reasoning, and be prepared to explain it. A well-reasoned exclusion is as valuable to a claim's credibility as a well-reasoned inclusion.

Educational material, not advice. Eligibility depends on the facts and the current rules in your jurisdiction.