Skip to content
Prime Partners, Chartered Accountants
Inside an Australian manufacturing facility
R&D for manufacturing

The work on the factory floor counts as research.

Australian manufacturers spend heavily on process improvement, new materials and production innovation, and a lot of them never claim, because the program sounds like it is meant for laboratories and software startups.

A company that systematically experiments with new processes, materials or designs to get past a technical problem is one of the strongest candidates there is. For a company turning over less than $20 million the offset is 43.5 per cent and refundable, so every $100,000 of qualifying spend returns $43,500 from the ATO, profitable or not. Prime Innovation, a specialist division of Prime Partners, works with manufacturers on identifying the activity, quantifying the spend and keeping records that hold up.

In short

If an engineer could predict the outcome, it is not R&D.

That one line settles most manufacturing questions. Work qualifies where it is experimental investigation carried out to generate new knowledge, and where the outcome cannot be known or determined in advance. Optimising within known limits, maintaining equipment, testing to a standard, and scaling up a proven process all sit outside it, however much engineering effort goes in.

What qualifies

Seven kinds of work that usually do.

New materials

Novel materials, composites or alloys where how they behave under production conditions is genuinely uncertain.

Process innovation

New manufacturing processes aimed at outcomes that cannot be reached by known methods.

Automation and robotics

Automated production systems where hitting the target speed, precision or reliability is technically uncertain.

Product development

New products where the design, the materials or the method of making them involves unresolved technical problems.

Quality improvement

Systematically finding the root cause of a defect and developing a solution that takes more than standard troubleshooting.

Sustainable manufacturing

New processes to cut environmental impact where whether the target can be met at all is an open question.

Tooling

New tooling, jigs or fixtures that need experimental work to reach the performance required.

What does not

Improvement is not the same as experiment.

Clearly outside

Routine quality control: standard testing, inspection and monitoring by established methods.
Optimising production within known ranges, to the manufacturer's specifications.
Maintenance and repair, routine servicing and breakdowns.
Scaling up production using a proven process with no new technical problem in it.

Also outside

Compliance testing against a regulatory standard by known methods.
Installing and running proven equipment according to the manufacturer's guidelines.
Minor product variations: colour, size or cosmetic changes with no experimentation behind them.

Examples

Four that would qualify.

A composite panel from waste

A building products manufacturer developing a panel from recycled plastic and natural fibres, testing fibre to plastic ratios, binding agents and manufacturing temperatures against structural strength, fire resistance and weathering.

Cutting sheet metal waste

A metal fabricator experimenting with new cutting pattern algorithms and alternative tooling geometries, to take waste from 18 per cent to under 8 while holding dimensional accuracy and line speed.

An automated packing line

A food manufacturer developing automated packaging for variable dimension products at 40 per cent faster rates. The uncertainty: whether vision-guided robotics can hit the picking accuracy and the speed at the same time.

A corrosion-resistant coating

An agricultural equipment maker developing a multi-layer coating system for harsh rural conditions, aimed at a fifteen year life under UV, chemical and mechanical stress.

A worked example

A mid-sized components manufacturer.

Forty five employees including six in engineering, aggregated turnover $12 million. The R&D is developing a high performance polymer component to replace a metal part.

CategoryTotal costEligibleWhy
Engineering team salaries$520,000$390,00075% on R&D, from timesheets
Production staff on trial runs$180,000$54,00030%, during the experimental runs
Raw materials$95,000$95,000Consumed in trials only
Prototype tooling$120,000$120,000Moulds for experimental components
External laboratory testing$45,000$45,000Mechanical and thermal testing
Equipment, decline in value$60,000$44,00073% apportioned to R&D
Total$1,020,000$748,000 

$748,000 at 43.5 per cent is a refundable offset of $325,380. For a capital intensive business putting money into the next generation of its products and processes, that is a real amount of cash coming back into the year it was spent.

Illustrative only. What a claim is worth depends on your facts and your apportionment.

Keeping records

The evidence has to be captured during the run.

Manufacturing R&D is physical, which means the proof exists for a few hours on the line and then it is gone. It cannot be reconstructed afterwards from memory and a delivery docket.

1
Trial records
Dated notebooks with the test parameters and the results, a trial production report for each experimental run, physical samples kept and labelled, and photographs or video of the setup.
2
Production data
Machine logs, CNC programs, PLC data and sensor readings. Quality results covering dimensions, material properties and failure analysis. Waste and yield compared between the experimental runs and the standard ones.
3
Project documentation
A technical specification setting out the problem and why existing solutions do not answer it, the hypothesis, the method and the success criteria, the results and conclusions from each phase, and minutes from the engineering meetings where the problems were argued about.
Related

The rest of the innovation picture.

Common questions

Questions, answered.

Does improving an existing process qualify?
Only where the improvement needs real experimentation. Adjusting machine speeds or temperatures within the manufacturer's specifications does not. Trying to reach performance beyond the established limits, and having to experiment systematically with new approaches, materials or configurations, can. The test is whether a competent manufacturing engineer could predict the outcome without running the experiment.
Can production trial runs be claimed?
Yes. Trials that form part of a systematic investigation into a technical uncertainty are eligible as core or supporting activity, including the materials consumed, machine time, labour and testing. You have to be able to separate experimental trials from routine production. Where product from an R&D trial is later sold or used, a feedstock adjustment adds an amount to your assessable income to take back part of the benefit on the inputs.
Are equipment purchases eligible?
The capital cost of buying equipment is generally not claimable. The decline in value of equipment used for R&D is, apportioned to the share of time it is used for R&D. Equipment used only for R&D gives 100 per cent of the depreciation. Equipment shared with production gives only the R&D portion.
How is this different from continuous improvement?
Lean, Six Sigma and Kaizen use established methods to optimise known processes, so the outcomes are predictable and the work is generally not R&D. R&D starts where continuous improvement runs out, at the point where the target cannot be reached by known methods and experimentation is needed to generate new knowledge. A continuous improvement project can cross into R&D when it meets a genuine technical barrier.
Can we claim while receiving a grant?
Yes, but not keep the full benefit of both on the same dollar. Where a grant covers expenditure you have claimed the offset on, a clawback adjustment adds an amount to your assessable income to take back the offset premium on the granted amount. Allocating costs carefully between grant funded and self funded work is what makes this manageable.
How far back can we claim?
Prior year returns can be amended within the ATO's standard amendment periods, generally two to four years. But the activities have to have been registered with AusIndustry within ten months of the end of that financial year. If registration was missed, that year cannot be recovered. Plenty of manufacturers have years of qualifying work that was never claimed, and it is worth looking back before assuming otherwise.

Walk us through the last thing that did not work first time.

That is usually where the claim is. If the answer was not obvious at the start and somebody had to try things to find it, there is a conversation worth having.

Contact Prime Innovation
Prime Insights

Clear thinking on the decisions that matter.

Once a month we send our latest thinking on tax, business, and the moments that shape a year. Practical, considered, and written for people who would rather understand the why than chase the headline.

Stay in the loop

Some of the biggest calls in a business and a life land between the appointments and the deadlines. Following along keeps you close to how we think, so when one of those moments turns up, you already know where to find us.

Trusted and accredited
★★★★★4.8 on Google from 70+ reviews100+ years combined experience
CPA AustraliaChartered Accountants ANZXero Platinum Partner