If your business claims the R&D Tax Incentive, the refund is probably one of the largest single cash receipts in your year. Which also makes it one of the largest single assumptions in your forecast.
Most of the forecasts we see make that assumption twice over. They assume the refund arrives roughly when it usually does, and they assume it arrives in full. Analysis referenced in Cut Through Venture’s Q2 2026 report suggests both assumptions are wrong often enough to matter.
In this article we’ll look at what the timing data shows – including why bigger claims wait longer and why the trend is going the wrong way – what happens to the amount you actually receive, and five changes you can make to your forecast so a late or light refund doesn’t take your plan with it.
A quick note on the source: the figures below come from Advanced’s analysis of almost 300 observed R&D refund events, published as part of the Cut Through Quarterly Q2 2026 report. It’s proprietary data that hasn’t been independently validated, and Advanced provides R&D advance funding, so it’s worth reading with that in mind. That said, the sample size is substantial and the pattern it describes matches what we see in practice.
The Median Hides the Tail
Across the dataset, the overall median was 13 days from the point the income tax return was processed to the refund arriving.
It’s worth reading that sentence carefully, because the measurement basis matters. The clock starts after processing – not when you lodge. The wait you actually experience, from lodgement to money in the bank, is longer than 13 days, and how much longer depends on ATO processing times you don’t control.
Then there’s the distribution:
| Share of refunds | Time after return processed |
|---|---|
| Median (50%) | 13 days |
| 25% | more than 25 days |
| 20% | more than 30 days |
| 10% | beyond 48 days |
A median of 13 days sounds manageable. A one-in-ten chance of waiting seven weeks is a completely different planning problem, particularly if that refund is funding payroll.
Bigger Claims Wait Longer
This is the finding most likely to affect the companies we work with, and it’s why the overall median is misleading if you’re a larger claimant.
Refunds below $250,000 had a nine-day median. Claims of $1 million or more had a median of 47 days, with 77% taking over 30 days.
The dataset is dominated by smaller claims, which is why the overall median sits so close to the small-claim figure. If your claim is substantial, 13 days isn’t the relevant benchmark. It’s closer to seven weeks.
There’s an uncomfortable logic to it, too. The companies with the largest R&D claims tend to be the companies with the largest payroll obligations, so the businesses that most need the cash to arrive on schedule are the ones waiting longest for it.
And It’s Getting Slower, Not Faster
Timing from prior years can’t be extrapolated forward. The dataset median rose from nine days in FY24 to 16 days in FY25.
If your forecast is built on how long your refund took two years ago, it’s built on a number that has since roughly doubled. Plan on the current environment, and plan on it not improving.
The Amount Isn’t What You Think Either
Timing is only half of it.
In 37% of the refund events analysed, the cash received was below the gross tax credit on the return. The median shortfall was 16.5%, reflecting ATO offsets applied against other liabilities – usually PAYG withholding or GST.
That’s the mechanism worth understanding. If you’ve got outstanding obligations to the ATO, your refund can be applied against them before anything reaches your account. Nothing has gone wrong when this happens; the system is working exactly as designed. But if you’re forecasting the gross entitlement, you’re forecasting a number that more than a third of the time won’t arrive in full.
If your refund is $800,000 and you’re planning around $800,000, a 16.5% offset is $132,000 of runway you thought you had.
What a Better Forecast Looks Like
Five changes, in rough order of impact.
1. Forecast on the 75th percentile, not the median.
The median is the wrong statistic for a planning assumption. You’re not trying to predict the typical outcome – you’re trying to build a plan that doesn’t break in an unfavourable one. Use a timing assumption you’d be comfortable being wrong about in only a quarter of cases.
2. Model net receipt, not gross entitlement.
Check your current ATO position – PAYG withholding, GST, anything outstanding – and model the refund net of what could reasonably be offset against it. If your position is clean, you’ll be pleasantly surprised. If it isn’t, far better to know now.
3. Stress-test your 13-week forecast for a 50-day delay.
Take your largest expected receipt, push it out by 50 days, and see what breaks. Payroll, super, rent, key supplier payments – if any of them fail in that scenario, you’ve identified the buffer you need to hold, and you’ve identified it while there’s still time to arrange one.
4. Keep fixed obligations independent of the refund.
Wages and super are fixed in amount and timing. An R&D refund is variable in both. Never let the first depend on the second. That’s doubly true now that payday super has removed the quarterly buffer that used to absorb this kind of timing risk.
5. Control the part you can control – lodgement.
You don’t control ATO processing times. You do control when you lodge, and how clean the claim is when it lands. Contemporaneous documentation, well-classified expenditure and early preparation all shorten the part of the timeline that’s genuinely yours. A claim that prompts questions is a claim that waits.
Where This Connects to Everything Else
Three things are true at once for Australian startups right now.
Equity capital has concentrated sharply, with a small number of large rounds taking most of the money and deal counts at their lowest in years, so the next round is a weaker base-case assumption than it used to be. Payday super has turned a quarterly cash obligation into a weekly or fortnightly one, removing the flexibility that used to absorb late receipts. And the R&D refund many companies rely on to bridge the gap arrives later, and sometimes lighter, than forecast.
Each of those is manageable on its own. It’s the combination that catches companies out, and it’s why the forecast matters more than usual this year.
What to Do Now
- Find your last three R&D refunds and work out the actual elapsed time from lodgement to cash received. Compare it to what your forecast assumed.
- Check your ATO position for anything that could be offset against the next refund.
- Re-forecast on the 75th percentile and net of offsets, then see whether the plan still holds up.
- If it doesn’t, look at advance funding options now – while you’re negotiating from a position of choice rather than need.
Want a Second Look at Your R&D Forecast?
Getting the claim right is one thing. Getting the cash flow around it right is another. If you’d like us to pressure-test how your R&D refund sits in your forecast, or talk through whether advance funding makes sense for your claim, it’s a much easier conversation to have now than in the week a refund fails to land. Get in touch with the Standard Ledger team today.
Disclaimer: This article is general in nature and does not constitute financial or tax advice. R&D Tax Incentive eligibility, refund timing and offset outcomes vary depending on your circumstances – speak to your Standard Ledger advisor for guidance specific to your situation. Third-party data referenced in this article has not been independently verified by Standard Ledger.
