General information only-not tax, financial, or legal advice.
This article explains common substantiation shortfalls the ATO and AusIndustry look for. Every business is different. Confirm what applies to your situation with a registered tax agent before you lodge. GrantsMAX prepares an evidence-backed pack; a registered tax agent reviews, refines, and lodges, and you own the claim.
Prerequisites
Before you start closing substantiation gaps, gather three things.
- Access to your accounting system-Xero, MYOB, QuickBooks, or similar. You will need the general ledger and transaction details that show R&D labour, contractor, and consumable costs. The GrantsMAX browser connector links read‑only to Xero, MYOB, QuickBooks, Microsoft 365, and Google Workspace so the platform can pull the right data.
- A project or activity list-even a rough list of the R&D projects you believe may be eligible. If you haven’t mapped them yet, start with the work that involved uncertainty, experimentation, or systematic trial‑and‑error.
- Your key contemporaneous records-emails, notebook scans, meeting agendas, Jira tickets, Slack threads, Teams logs. You don’t need a perfect archive; you need enough to show what was done, when, and why.
With those in hand, the sequence below walks through the most common gaps and the practical fixes that make a claim defensible.
Why R&D substantiation gaps matter
The R&D Tax Incentive is a self‑assessment program jointly administered by the ATO and AusIndustry (Department of Industry, Science and Resources). A business registers its R&D activities with AusIndustry and then claims the R&D tax offset in its tax return. The ATO expects the entity to keep records that substantiate both the activities and the expenditure.
When a review or audit occurs, the ATO will ask for contemporaneous evidence that:
- the activities registered genuinely constitute core or supporting R&D activities as defined by the Income Tax Assessment Act 1997,
- the expenditure claimed is directly related to those activities, and
- the business has a demonstrable hypothesis‑testing or systematic experimental process.
A substantiation gap is not just a piece of missing paperwork. It can lead to a disallowed claim, an amendment, penalties, or a reputation risk for the accountant or tax agent. The most frequent gaps are well‑known to the ATO and AusIndustry, and they surface again and again in compliance‑focused publications. Addressing them before you lodge is far simpler than reconstructing evidence after the fact.
Pro tip: Even if you’re not under review right now, building an annual evidence discipline means next year’s claim is ready faster. The Annual Refresh & Accountant Channel helps you carry forward project structures and evidence patterns so you’re not starting from scratch each year.
Gap 1: Missing or inadequate timesheets
Why this gap appears
For many R&D‑active firms, especially software, manufacturing, and agtech, the bulk of the claim is labour. The ATO expects a reasonable record of the time spent on R&D activities. Yet many businesses rely on after‑the‑fact estimates, a single line in a project plan, or email threads that mention work was done but don’t capture hours.
In an ATO review of the R&D Tax Incentive published on ato.gov.au, the agency highlighted that “contemporaneous records of the time spent on R&D activities” were a common deficiency. It also noted that estimates made many months after the fact rarely meet the standard required.
Practical fix
- Set up a timesheet category or tag in your existing system (Xero Projects, MYOB, or even a shared spreadsheet). Every person doing R&D logs their time against a specific project code, with a two‑line description of the work and the hypothesis being tested.
- Review weekly. A five‑minute Friday check ensures the records are fresh and complete. This is far easier than trying to reconstruct 300 hours a month later.
- Tie the timesheet data to payroll. When you extract reports from your accounting system via the GrantsMAX browser connector, the labour costs will already match the timesheets. The AI Application Pack Drafting feature then pulls those figures into a cost-structure ready for your accountant to review.
- For key experiments, attach a supporting note. Even a single paragraph documenting what was attempted, the uncertainty, and the result can turn a timesheet line into a solid activity narrative.
Warning: A timesheet entry that says “R&D work” or “Project Alpha” without any description of the systematic experimentation will rarely satisfy an ATO reviewer. The ATO consistently states that records must show the link to the R&D activity for which you are claiming.
Gap 2: Vague activity descriptions that don’t link to core R&D
The problem
AusIndustry requires registration of “core R&D activities” and “supporting R&D activities”. Core activities must involve an experiment whose outcome cannot be known in advance and that is conducted for the purpose of generating new knowledge. The registration form asks for a description of the systematic progression of work.
Too many registrations use generic language: “developing new software features”, “improving manufacturing yield”, or “researching new product formulations”. These descriptions don’t make the hypothesis, the uncertainty, or the experimental methodology explicit.
Practical fix
- Start each description with the hypothesis. E.g., “We hypothesised that by adjusting spore concentration in fermentation vat A we could increase yield by X% without affecting product stability; the outcome was uncertain because the organism’s metabolic behaviour under the new media had not been tested.”
- Include the experimental approach. Mention the variables changed, the control group, the number of iterations, and the method of measurement.
- State the outcome specifically. Even a negative result qualifies if the experiment was systematic and documented.
When GrantsMAX drafts the activity narratives, it surfaces these elements from your data and from your contemporaneous emails and notes. The pack is then reviewed by your registered accountant. You can read more about how the platform maps activities to the statutory definitions in our guide What is the R&D Tax Incentive? A plain-English guide for Australian businesses.
Gap 3: No contemporaneous records beyond financial data
The gap
The ATO and AusIndustry both point to the importance of records created at the time the R&D happened. Emails, meeting notes, lab books, Jira tickets, Slack messages, design review minutes, these often exist but are scattered. In an audit, a business may struggle to present a cohesive narrative.
The OECD’s work on Business R&D and Innovation underscores that robust evidence trails are critical for policy evaluation globally. International datasets, such as those from the NSF National Center for Science and Engineering Statistics, consistently show that firms with systematic record‑keeping report higher confidence in their R&D claims.
Practical fix
- Choose a lightweight evidence‑capture system. For software teams, link GitHub commits to Jira issues. For manufacturers, photograph test rigs and date‑stamp them. For agtech, keep field‑trial logs.
- Build an evidence index. The Audit-Ready Evidence Trail feature in GrantsMAX creates a cross‑reference that ties each activity and cost line to its source document, email, invoice, or timesheet, so your accountant has a clear picture.
- Don’t over‑engineer it. A dated email with a brief experiment summary, a snapshot of a whiteboard, or a three‑bullet Slack update can be sufficient, provided it captures the hypothesis and observation.
For broader context on how governments view data gaps, the U.S. Federal Register published a Request for Information on Identifying Critical Data Gaps that highlights the same themes: without contemporaneous records, the underlying activity is hard to verify.
Gap 4: Poor cost allocation and missing nexus
The issue
You may have timesheets and invoices, but if you can’t demonstrate why a cost was incurred directly on R&D activities, the ATO may disallow part of the claim. Common errors:
- Claiming 100% of a senior engineer’s salary when only 60% of their time was on R&D.
- Including consumables that were also used in routine production.
- Allocating overhead without a consistent, reasonable basis.
Practical fix
- Segregate R&D costs in your chart of accounts. Use a distinct tracking category or project code in Xero/MYOB/QuickBooks for each R&D project. The GrantsMAX browser connector can pull these category balances, making cost mapping straightforward.
- Document the allocation methodology. For example, a machine’s depreciation might be apportioned by hours used on R&D vs. production; keep a log of those hours.
- Tie consumables to specific experiments. If you order chemicals, prototype parts, or cloud computing credits, record which R&D activity they supported and the experiment they fed into.
- Use the ATO’s R&D tax incentive schedule. The schedule asks for a breakdown of expenditure. When you use AI Application Pack Drafting, the system pulls expense data directly from your accounting software and maps it to the schedule structure, so your accountant can see the nexus and adjust if necessary.
Pro tip: The ATO has indicated via its public guidance that where a cost is shared, the business must be able to explain the apportionment basis. Keep that explanation in a short note attached to the cost line.
Gap 5: Failing to capture hypothesis testing and uncertainty
The gap
Core R&D activities require a systematic progression of work that is based on principles of established science and proceeds from hypothesis to experiment, observation, and evaluation. The most common oversight is describing only the technical task, not the unknown outcome.
Research on research funding reproducibility, such as a recent article in Science, highlights how rigorous documentation of hypothesis and method strengthens the evidence chain. The same principle applies in the tax context: if you can’t show the uncertainty, the activity may look like routine development.
Practical fix
- Document the knowledge gap explicitly. What did you not know before the experiment? Why could the outcome not be determined in advance?
- Describe the systematic approach: variables tested, controls, measurement methods, iterations.
- Record both successes and failures. The program is agnostic to commercial success; a well‑designed experiment that fails to achieve its goal is still eligible if it meets the definition.
- Use structured templates. GrantsMAX provides a narrative framework that prompts the user for hypothesis, method, observation, and evaluation. The output is a draft that a registered accountant reviews and refines in the Accountant Review & Lodge Workflow.
Gap 6: Misclassifying supporting and core activities
The distinction
Supporting R&D activities are those directly related to core R&D activities, but they must be undertaken for the dominant purpose of supporting that core R&D. Common mistakes:
- Claiming general market research as supporting R&D.
- Claiming routine data entry or admin as supporting.
- Claiming product management tasks without a direct, documented tie to a specific core experiment.
Fix
- Map each supporting activity to a specific core activity. For example, “preparing bacterial cultures for the fermentation experiment dated 12 March” is a supporting activity if it directly enabled that experiment.
- Ensure all supporting activities have a clear “dominant purpose” link. The ATO’s guidance says the activity must be produced for the dominant purpose of supporting core R&D.
- Review the registration annually. The Eligibility Assessment & Risk Flags feature flags areas where the classification may be ambiguous, so your accountant can address them before lodgement.
Gap 7: Losing project management and supervision evidence
Why it’s a gap
Project management, supervision, and quality assurance can be supporting R&D activities if they directly support core experiments. But businesses often forget to capture records of scrum ceremonies, design reviews, and lab‑team meetings. The NIST Manufacturing Extension Partnership notes that even mature manufacturers struggle to document the operational context of innovation, which can leave their claims exposed.
Fix
- Save agendas and minutes. A weekly R&D project review that lists hypotheses being tested, adjustments made, and next steps is a powerful contemporaneous record.
- Include Jira/Sprint reports. For software teams, the sprint retrospective can show systematic progress and evidence of iteration.
- Tie supervision records to timesheets. If a senior scientist supervised an experiment, their timesheet entry should reference the experiment and the supervising role.
How GrantsMAX fits in
GrantsMAX is not a lodgement service. It is an AI grant agent that connects read‑only to your accounting data (Xero, MYOB, QuickBooks, Microsoft 365, Google Workspace) via its browser connector. It then discovers the government grants and R&D tax incentives you may be eligible for and prepares a complete, evidence‑backed application pack. A registered accountant or tax agent reviews, refines, and lodges the claim; the business owns it. You can see the workflow in the Accountant Review & Lodge Workflow page.
The platform covers the R&D Tax Incentive, EMDG export grants, and state innovation grants. For technology companies, the GrantsMAX for technology companies page explains how software and product engineering may be eligible and where the narrative often fails. For manufacturers, GrantsMAX for manufacturers shows how process improvement and automation can be framed.
By integrating with your live data, GrantsMAX builds an Audit-Ready Evidence Trail and refreshes claims annually through the Annual Refresh & Accountant Channel. Accountants and bookkeepers can white‑label the workflow, making it easier to substantiate claims across a client base.
Pro tips for a defensible claim
- Start early. The best time to capture evidence is when the work is happening. Even a five‑minute daily log can save weeks of reconstruction later.
- Use the ATO’s R&D tax incentive schedule as a checklist. If a cost category appears in the schedule, ask “Do I have contemporaneous records for this line?”
- Involve your tax agent long before lodgement. The Accountant Review & Lodge Workflow is built precisely for this early collaboration, so the agent flags gaps while there’s still time to fill them.
- Review the business.gov.au guidance annually. The AusIndustry website publishes updated examples of core and supporting activities. The Grants & R&D Discovery and Matching feature scans these changes so you stay informed.
- Don’t let “perfect” be the enemy of “good.” A regular, if imperfect, contemporaneous record is far better than a polished document created after the fact. The UK’s Research and development statistics collection shows that even large firms can struggle with tidy evidence; what matters is consistency.
Summary and key takeaways
- The ATO and AusIndustry look for contemporaneous records that show a systematic, hypothesis‑based experimental process and a clear nexus between activity and cost.
- The most common substantiation gaps are absent timesheets, vague activity descriptions, missing contemporaneous evidence, poor cost allocation, and misclassification of supporting activities.
- Closing these gaps is a matter of small, regular habits: daily time logging, brief experiment notes, and consistent cost tagging.
- Tools like GrantsMAX can help surface gaps from your live data and draft a pack that a registered accountant reviews and lodges, but the business must maintain the underlying records.
- Because rules and ATO guidance change, always confirm the current requirements with a registered tax agent or directly from ato.gov.au and business.gov.au.
If you want a faster, more rigorous path from raw data to a review‑ready R&D claim, join the GrantsMAX waitlist. We’ll help you connect your accounting system and start building an evidence trail that gives your accountant what they need to lodge with confidence, without the last‑minute scramble.
Join the waitlist at https://www.grantsmax.com.