Is a pilot plant eligible for the R&D Tax Incentive?

Pilot plants and the R&D Tax Incentive, from experiment to asset

Sometimes eligible

Sometimes. Running a pilot plant can be core R&D when each campaign tests a hypothesis about behaviour bench work could not predict. Building it is usually supporting, unless the configuration itself is tested, and is claimed through decline in value. Sold or used output can trigger a feedstock adjustment; commissioning and running at known settings are not R&D.

Reviewed by George WalchFacts last verified 24 October 2026

A sorbent unit on a real kiln gas stream

An illustrative company has a solid sorbent that captures carbon dioxide well from clean gas in the laboratory. Cement kiln flue gas carries sulfur dioxide and dust, and its search found no data on how this sorbent's capacity holds over repeated cycles in that gas. It builds a small pilot unit on a slip stream and runs campaigns varying regeneration temperature, each against a recorded capacity loss threshold. Those campaigns are core; building the unit is supporting.

Proving a commercial electrolyser on site water

A company installs a commercially available electrolyser at its site, treats locally sourced water to the maker's specification and runs it to confirm the hydrogen purity it will achieve, then makes incremental adjustments to find settings for its duty cycle. AusIndustry's energy guide treats the purity check as neither core nor supporting, because the maker's specifications already indicated the outcome, and says incremental adjustments to find operating settings are unlikely to be core.

The evidence that supports it

The bench results and the search showing why behaviour at pilot scale could not be predicted from them

A campaign plan for each run stating the hypothesis, variables and threshold

Operating logs marked to show test campaigns separately from commissioning and idle periods

The fixed asset register entry for the plant, with the date it was installed ready for use and its R&D use share

Records of every output stream (sold, used on site, stored or disposed of) for the feedstock calculation

Grant agreements and milestone reports for any government funding of the plant

A pilot plant sits between the laboratory and the first commercial unit. It is where many process companies do their most important experiments, and also where claims most often blur experiment, construction and production. The R&D Tax Incentive treats each of those differently, and the ATO adds rules of its own once the plant exists as an asset.

Operating the plant: experiment or operation?

A campaign on a pilot plant is core R&D when it tests a hypothesis about behaviour the bench could not predict: impurities in a real feed, heat removal at a larger throughput, catalyst or sorbent life under true cycling. The general rule for scale-up applies here as well: the company must be able to say why the smaller-scale results could not answer the question.

AusIndustry's energy guide marks the other side. Commissioning tests that confirm performance against known requirements are routine testing, and incremental adjustments to find operating settings are unlikely to qualify, because the relationship between settings and results is already known. Its electrolyser example failed for that reason. Operating hours between test campaigns are not R&D either. Feed materials, energy, consumables and operator time used during a test campaign are expenditure on the core activity, so log campaign start and stop times and apportion running costs to them rather than to the plant's whole operating year.

Building the plant: supporting, and claimed as an asset

Designing and building a pilot plant enables experiments; it is rarely the experiment. It is usually supporting, unless a particular configuration is itself the variant under test, such as a reactor geometry whose heat removal is the unknown. How it is claimed matters more than its label:

Expenditure included in the cost of a depreciating asset is not a notional R&D deduction. The plant's decline in value is claimed instead, to the extent it is used for R&D.
Expenditure to acquire or construct a building is excluded, but the ATO notes that a building which is itself plant or a depreciating asset may attract R&D decline in value.
Salary expenditure incurred on constructing depreciating assets is excluded under s 355-225.

What the plant produces: feedstock

The ATO's feedstock rules apply when R&D transforms or processes goods into tangible products that are supplied to others or applied to the company's own use. The adjustment covers the feedstock inputs, energy input directly into the transformation, and the decline in value of assets used to acquire or produce the inputs. It is worked out on the lesser of the feedstock expenditure and the feedstock revenue, and recovers the incentive part of the offset by adding an amount to assessable income.

When the plant is sold or converted

When the company stops holding an R&D asset, the ATO applies a balancing adjustment, then claws back the incentive component on any gain or allows a catch-up deduction on a loss. Grants for the same R&D are handled separately: one expected when the money was spent can engage the at-risk rule, and one that was not can trigger a clawback. Which applies depends on the grant's terms and timing.

The physics of scale is covered on manufacturing scale-up, and smaller rigs on prototypes and tooling. Energy sector context is on climate and energy software.

Frequently asked questions

Is a demonstration plant treated the same as a pilot plant?

Not usually. A pilot plant exists to resolve unknowns. A demonstration plant usually exists to show that a known process works at scale to investors, customers or financiers, and its dominant purpose is commercial. Any genuine experiments run on it are assessed for the test periods only.

The pilot plant will become our first production unit. What changes?

From the point the plant's use turns to production, its decline in value is no longer for R&D, and production runs are not R&D activities. The core activity also ends once the process is predictable. Record the date the plant's use changed.

We received a grant to build the plant. Can we still claim it?

Possibly, but the grant reduces the benefit in one of two ways, depending on the facts. If the company could reasonably expect the grant when it incurred the expenditure, the at-risk rule may reduce the notional deduction itself. Otherwise a grant or recoupment for the same R&D triggers a clawback adjustment. Keep the grant agreement and its milestones with the claim records so the company's accountant can apply the right rule.

The R&D Tax Incentive is a self-assessment program. This page is general information, not tax, legal or financial advice; eligibility depends on your circumstances and you should seek independent advice for them.