R&D Tax Incentive

The core test is whether the project involves genuine experimental activities — where the technical outcome cannot be determined in advance and requires a systematic investigation to resolve. Custom software development often qualifies, particularly where novel approaches or new integrations are involved. We assess eligibility at scoping and give you a clear view before you commit to the process.

Companies with aggregated turnover under $20M can access a 43.5% refundable offset, meaning it can reduce your tax liability below zero and produce a cash refund. Companies above that threshold access a 38.5% non-refundable offset, which reduces tax owed but cannot produce a refund. Both are meaningful. Your tax specialist will confirm which applies to your entity.

Registration with AusIndustry must be completed within ten months of the end of the income year in which the R&D activities were conducted. That sounds like plenty of time, but documentation gathered throughout the project is significantly stronger than documentation assembled from memory at registration. Starting at scoping makes the claim more defensible and more complete.

We work with a specialist R&D tax advisory firm experienced in ATO-compliant claims for software and technology projects. We make the introduction as part of the project onboarding process. All tax advice and lodgement is managed by them, not by CodeFish.

R&D Tax Incentive support is structured into Amplify projects where eligible activities are identified at scoping. It is not available as a standalone service independent of a CodeFish project engagement.

You need records made while the work happens that show what you were trying to find out, what you tried, and what you learned. That means a written hypothesis for each experimental activity, the tests you ran, the results, and the decisions that followed. It also means timesheets or other records linking staff and contractor time to those activities, and the invoices behind the spend. On a CodeFish project, much of this comes from work we already produce: tickets, technical notes, test results and the code history. Your tax adviser will tell you what else they need and in what form, so it is captured as you go rather than rebuilt at year end.

If AusIndustry or the ATO reviews your claim, you answer with the records you kept during the project, and your tax adviser manages the response. Reviews look at whether the registered activities were genuine R&D and whether the expenditure claimed relates to them. Where a review finds that an activity or cost was not eligible, the offset can be reduced and any overpaid amount may have to be repaid. That is why the claim is framed at scoping and documented as the work happens. A claim built on contemporaneous records is far easier to defend than one assembled from memory after the financial year ends.

No. Routine software development does not qualify on its own, even when it is new to your business. Building screens, setting up user logins, writing standard reports or connecting systems with a well-documented method are generally not R&D, because a competent developer can work out the answer in advance. What can qualify is the part of a project where the outcome is genuinely unknown and the only way to find out is to test a hypothesis and measure the result. Most builds are a mix of both. We help you separate the two at scoping, and your adviser confirms where the line sits for your project.

The benefit lands after you lodge your company tax return for the income year in which the R&D was done, once the ATO processes it. Registration with AusIndustry comes first, within ten months of the end of that income year, and the R&D schedule is lodged with the return. For a company that is eligible for the refundable offset and has no tax to pay, the result is a cash refund. For a company with tax to pay, the offset reduces what is owed. The exact timing depends on when you register and lodge, so your adviser will tell you what to expect for your year.

A failed experiment does not by itself rule out an R&D Tax Incentive claim, because the incentive is aimed at work whose outcome could not be known in advance. If you tested a hypothesis properly, recorded what you tried and learned that the approach does not work, that work can still be eligible R&D. What matters is the systematic method and the records, not whether the result was the one you hoped for. On a CodeFish project, a failed approach is written up the same way as a successful one, with the hypothesis, the tests and the decision that followed. Your tax adviser confirms whether the activity qualifies for your claim.

Most of the R&D Tax Incentive paperwork comes from records the project already produces, so the extra load on your team is usually small. On a CodeFish project, tickets, technical notes, test results and the code history capture the experimental work as it happens. Your team's main jobs are keeping time records that link staff to those activities, holding on to the invoices, and answering your adviser's questions at year end. Your tax adviser handles the registration with AusIndustry and prepares the schedule that goes with your company tax return. If the paperwork looks likely to cost more than the benefit, we will tell you at scoping rather than let you find out later.

Planning for an R&D Tax Incentive claim changes how the experimental parts of the project are recorded, not what you are building or what you pay us. The build is scoped and quoted on what the software needs to do, whether or not a claim is made. Where part of the work is genuinely experimental, we frame it as a hypothesis, plan the tests and record the results as we go. That adds structure to work that would happen anyway, and it gives you better evidence of what was tried and why. We never shape a project around the offset at the expense of the outcome you need.

If part of the work turns out not to be eligible, it simply stays out of the claim and is treated as an ordinary business cost. Eligibility is never fully known at the start, because some problems that look experimental are solved quickly with a known method. That is why we separate likely R&D from routine development at scoping and revisit the split as the build progresses. Your tax adviser decides what goes into the claim based on the records, and leaving out doubtful items is the safer choice. A smaller claim that holds up is worth more to you than a larger one that has to be repaid after a review.