Australia is the only one of the seven countries whose law splits R&D into two kinds of activity and asks the claimant to register each separately. Core R&D activities are experimental activities whose outcome cannot be known in advance from current knowledge. They follow a systematic progression of work, from hypothesis to experiment, observation and evaluation, leading to logical conclusions, and are conducted to generate new knowledge. Supporting R&D activities are directly related to a core activity, and some must also be undertaken for the dominant purpose of supporting it. Getting the split right is most of the work of an Australian claim.
Proposed changes from 1 July 2028, not yet law. Proposed from 1 July 2028, announced in the 2026-27 Budget (12 May 2026) and not yet legislated: a 4.5 percentage point increase in offset rates for core R&D; removal of supporting R&D activities from eligibility; intensity threshold reduced from 2 percent to 1.5 percent; refundable offset turnover threshold raised to A$50 million; minimum notional deductions raised from A$20,000 to A$50,000; maximum expenditure threshold raised from A$150 million to A$200 million; companies more than 10 years old no longer eligible for the refundable offset. Current rules apply until then.
What it is
| Element | Core R&D activity | Supporting R&D activity |
|---|---|---|
| Definition | An experimental activity whose outcome cannot be known in advance, determined by a systematic progression of work based on established science, conducted to generate new knowledge | An activity directly related to a core R&D activity |
| Statute | Section 355-25 | Section 355-30 |
| Extra test | Must not be an excluded activity | Dominant purpose test where the activity produces goods or services, is directly related to producing them, or is an excluded activity |
| Needs a hypothesis | Yes | No, but needs a named core activity it supports |
| Can be an excluded activity | No | Yes, if the dominant purpose test is met |
| Examples | Trialling an alloy at different heat-treatment schedules to find whether a hardness target can be met | Making the test coupons; calibrating the test equipment; a literature review scoping the hypothesis |
Who it applies to
Every registrant. There is no claim without at least one core activity: supporting activities exist only in relation to a core activity, and a registration of supporting activities alone is not eligible. A multi-year project can have core activities that start and stop in different years, and each year's registration lists the activities conducted in that year.
The dominant purpose test is where the split bites. It applies to a supporting activity that produces goods or services, or is directly related to producing them, or is one of the activities excluded from being core. Such an activity qualifies only if the dominant purpose for undertaking it was to support the core activity. A production run that doubles as a trial fails the test if its main purpose was to fill orders; a run made chiefly to test a hypothesis, with the output sold incidentally, passes, subject to the feedstock adjustment.
How it works
- Start from the technical unknowns, not the project. List the questions the team could not answer from existing knowledge. Each is a candidate core activity.
- Write the hypothesis and the experiment for each. If you cannot state what was proposed and how it was tested, it is not a core activity; it may be development.
- Attach the supporting activities. For each core activity, list the work done to enable or complete it: building rigs and prototypes, preparing samples, collecting data, calibration, literature reviews.
- Apply the dominant purpose test to any supporting activity that produces goods or services or is otherwise excluded, and record the reasoning.
- Set the boundaries. A core activity ends when the hypothesis is confirmed or rejected. Work after that point, such as scaling production or adding features, is not part of it.
- Register and keep records by activity, so that the department can examine each one on its own.
The department's Guide to Interpretation stresses that the systematic progression of work does not have to be elaborate, and that negative results are results: an experiment that disproves the hypothesis is a core activity. It also warns that describing a project in terms of its business goal, such as "develop a new app", does not describe a core activity.
Worked example
Illustrative example. Murray Composites Pty Ltd, a hypothetical company, wants to make a lighter irrigation pipe from a recycled plastic blend. The project as the business sees it has one goal. Registered as activities:
| Activity | Type | Why |
|---|---|---|
| Trialling blend ratios and extrusion temperatures to find whether burst pressure of 12 bar can be reached with at least 60% recycled content | Core | Unknown outcome, hypothesis tested by experiment |
| Testing UV additives to find whether field life of 10 years can be achieved with the recycled blend | Core | Second independent unknown |
| Modifying the extrusion die to run the trial blends | Supporting | Directly related; not production |
| Trial production runs of 2 km of pipe for burst and UV testing | Supporting, dominant purpose test | Produces goods; run chiefly to test, with offcuts sold as scrap, so passes |
| Grower survey on pipe diameters | Not eligible | Market research, excluded, and not supporting an experiment |
| Applying for the Australian Standard certification mark | Not eligible | Compliance with standards, excluded; not for the dominant purpose of supporting the experiments |
Murray registers two core and two supporting activities. The pipe sold as scrap brings the feedstock adjustment into play on the trial runs.
Common questions
Can a whole project be one core activity? Only if the whole project is one experiment. Usually a project contains several core activities and a larger number of supporting ones.
Does the experiment have to be in a laboratory? No. Field trials, pilot plants and production-line tests are experiments if the outcome was unknown and the work was systematic.
What if we were confident it would work? If a competent professional could have known the outcome in advance, it is not a core activity, however much work it took.
Can an excluded activity ever be claimed? As a supporting activity, if its dominant purpose was to support a core activity. Never as a core activity.
Related pages
Official sources
Educational content only, not tax advice. See the disclaimer.