Quick answer: Dividing experimental work between an internal engineering team, a university and an external specialist does not, by itself, weaken the research. The methodological risk arises when each team changes the conditions, measurements or interpretation rules independently. If the results are intended to answer the same technical question, the experimental protocol needs enough shared structure for the observations to remain comparable. For R&DTI purposes, however, methodological consistency does not by itself determine which entity can claim the activities or expenditure; those questions are assessed separately. You self-assess.
11 September 2026 — this article describes the current rules. The 2026–27 Federal Budget announced proposed R&DTI reforms for income years starting on or after 1 July 2028. Until any amendments take effect, the R&DTI continues to be administered under the current legislation.
Collaboration Changes the Experimental Environment, Not Just the Project Team
A technically uncertain project often reaches a point where one team cannot sensibly do everything. The company may understand the operating environment. A university group may have specialist modelling capability. An external laboratory may have equipment the company cannot reproduce internally. A software or engineering specialist may be responsible for another part of the experimental system.
That division of capability can be useful. Australian programs such as the Cooperative Research Centres Program and ARC Linkage are themselves structured around collaboration between researchers, industry and research end-users.
But collaboration creates a methodological problem that is easy to underestimate. Imagine three teams investigating the same question: whether a proposed control method can maintain a process within a defined operating range under changing conditions:
• Internal team: Tests on full-scale production equipment.
• University team: Tests a computational or physical model under controlled laboratory conditions.
• Specialist supplier: Tests an individual component on an isolated bench rig.
All three may describe their work as testing the same hypothesis. They may not actually be producing comparable evidence. One team may measure performance over ten minutes while another measures an instantaneous peak. One may exclude warm-up periods while another includes them. Different sensor calibrations, software versions or environmental conditions may change the result. Even the definition of “failure” may differ between sites.
At that point, the difficulty is no longer simply coordinating three organisations. It is preserving one research question across three experimental environments.
Agree What Must Remain Common Before the Work Separates
Australian Government guidance describes a systematic progression of work as moving through a hypothesis, experiment, observation, evaluation and logical conclusions. It also states that records of an experiment should explain how it is conducted or planned, including, for example, the parameters that are varied, held constant, observed or measured.
That becomes especially important when different teams conduct different parts of the investigation. Not every detail needs to be identical. In many real projects, it cannot be. A laboratory rig will not reproduce a production line perfectly, and a simulation may deliberately isolate conditions that cannot be isolated in the field.
The useful question is:
Which differences are part of the experiment, and which differences would make the results impossible to interpret together?
Before work is distributed, the teams should usually agree on a common experimental spine. This is a practical research-design discipline rather than a prescribed R&DTI template. That may include:
• The technical question being investigated
• The hypothesis or linked hypotheses
• The dependent measure that ultimately answers the question
• The definitions of important operating conditions
• The control or baseline against which results are compared
• The measurement method and units
• The treatment of missing or invalid observations
• The versions of relevant hardware, software, datasets or models
• The conditions that must be recorded when they cannot be held constant
• The result that would support, weaken or require revision of the hypothesis
This does not turn every site into a copy of the others. Instead, it gives the research team a way to distinguish a genuine experimental result from a difference caused by how the experiment was run.
For example, if the laboratory sees an improvement and the production site does not, that disagreement may be useful evidence — but only if the team can identify whether operating scale, material variation or another recorded condition explains the difference. If the two teams used different definitions of the performance measure, the disagreement tells you much less.
Treat Deviations as Evidence, Not Administration
Multi-party research rarely follows the original protocol exactly. A supplier may change a component. A university researcher may discover that a measurement method lacks sufficient resolution. The production team may be unable to hold an environmental variable constant. A software update may alter the behaviour of the system halfway through the trials.
Treating those differences as administrative noise in order to produce a cleaner dataset is the wrong response. The deviation itself belongs in the research record. Government R&DTI record-keeping guidance emphasises records created at the time activities are conducted, rather than reconstructed retrospectively. Clear records help establish what work occurred and provide the context needed to interpret the activity and its results.
For collaborative experiments, a useful deviation record can be short:
What changed?
Identify the equipment, method, parameter, dataset, environment or procedure.
Why did it change?
Was the original method impossible, unsafe, unreliable or shown to be inadequate?
When did it change?
Record which experimental runs occurred before and after it.
What does the change affect?
Can earlier and later results still be compared directly, or should they be analysed as separate experimental conditions?
Who made the technical decision?
Record the reasoning, not merely the approval.
This matters because a research result is rarely just a number. Its meaning depends on the circumstances under which the number was produced. A dataset showing ten successful trials across three organisations may appear more persuasive than one produced by a single team. But if the teams used materially different protocols and those differences were not recorded, combining the results may create confidence that the evidence does not support.
One Project Can Contain Several Linked Experiments
Shared governance does not mean forcing every collaborator into one identical experiment. Sometimes the better design is to recognise that the project contains several linked experimental questions:
• The university may investigate the underlying mechanism.
• The specialist laboratory may test whether a component exhibits the predicted behaviour.
• The company's engineering team may test whether that behaviour persists under the operating conditions that matter commercially.
Those experiments can contribute to one broader technical decision without pretending that they are interchangeable. Collaborative Australian research programs also illustrate that technically related work can be distributed across different organisations. The CRC Program supports industry-led collaboration between industry and research organisations, while ARC Linkage Projects foster research alliances between higher education organisations and research end-users. Neither program, however, determines how a particular activity should be treated under the R&DTI.
The methodological task is to make the interfaces explicit:
• What result from Experiment A determines the design of Experiment B?
• Which assumption from the laboratory must be tested again at operating scale?
• If two results conflict, who is responsible for deciding whether the hypothesis changes or another experiment is required?
• Which record links that decision back to the observations that caused it?
Without those links, a collaborative project can produce several technically competent reports that never quite add up to one answer. With them, the differences between teams become part of the research design rather than noise around it.
Collaboration and R&DTI Treatment Are Separate Questions
A coherent experimental protocol does not, by itself, determine which organisation may claim the work under the R&D Tax Incentive. Section 355-210 of the Income Tax Assessment Act 1997 separately considers whether an R&D activity is conducted for the R&D entity and whether it is conducted, to a significant extent, for another entity. Activities conducted outside Australia are also subject to separate requirements. The contractual structure, control of the work, ownership or use of the results, financial risk and location of the activities may therefore matter independently of whether the experimental evidence is scientifically comparable.
Conclusion
Collaborative research does not require every team to run an identical experiment. It does require the differences between experimental environments to be deliberate, visible and interpretable.
A shared technical question, agreed measures and controls, clear deviation records and explicit links between experiments allow evidence from different teams to contribute to one reasoned conclusion. Where those links are missing, more collaborators can produce more data without necessarily producing a stronger answer.
Sources & Further Reading
Australian Government, business.gov.au — Conducting core R&D activities for the R&D Tax Incentive
Australian Government, business.gov.au — Record keeping for the R&D Tax Incentive
Australian Government, business.gov.au — Getting help from a Research Service Provider
Department of Industry, Science and Resources — Cooperative Research Centres Program
Australian Research Council — Linkage Projects
Federal Register of Legislation — Income Tax Assessment Act 1997, s 355-210 (Conditions for R&D activities)
Talk to Ignition Research if a technically uncertain project involves internal teams, laboratories, universities or external specialists and the experimental work needs to remain comparable across them. As a Registered Research Service Provider, Ignition Research assists businesses with research question definition, experimental design, technical investigations and contemporaneous supporting records within its registered research fields.
RSP registration confirms that the provider has met the relevant registration requirements for its registered research fields. It does not constitute government endorsement of a project or guarantee R&DTI eligibility.
We do not determine R&DTI eligibility or provide tax advice. The R&D Tax Incentive is a self-assessment program, and companies remain responsible for assessing their activities, expenditure and circumstances and obtaining appropriate tax advice.
This article provides general information only and is not tax, legal or financial advice.
Working on a project like this?
Tell us what you're building and the part you're not sure will work. A researcher from our Adelaide team (registered Research Service Provider RSP000047) reads every enquiry and replies within two business days. No documents needed.
