In 17 Minutes, We Found ~$1 Million in R&D Credits They Didn't Know Existed

A founder reached us through a referral in early 2026. We jumped on a call. They connected their Xero file while we were talking, and seventeen minutes later we had a working estimate of roughly $1 million in R&D credits they could potentially access over the next three to four years — three years of RDLTC and three years of RDTI.

They'd been running qualifying R&D for two years and had no idea.

The Good Intentions Problem

Here's what we found when we pulled up their accounts: just over $1 million in software development costs sitting on the balance sheet as an intangible asset. Capitalised.

This is one of the most common patterns we see in early-stage tech companies. An accountant — acting reasonably — advises the client to capitalise software development spend rather than expensing it. The logic is sympathetic: it smooths the P&L, reduces the reported loss, makes the company look less like it's burning cash. Good intentions.

The problem is that it creates two separate issues.

The first is an accounting one. IAS 38 — the standard governing intangible assets — requires that for development costs to be capitalised, the entity must be able to reliably estimate the asset's useful life and demonstrate that the project is technically and commercially feasible. For most startups, both of those assumptions are genuinely questionable. If your product is still finding product-market fit, if you're burning cash and don't yet know the timing or size of your next raise, if the roadmap changes quarterly — you are arguably not in a position to assert that this software has a determinable useful life. Capitalising under those conditions isn't conservative; it may simply be non-compliant.

The second issue is the R&D one. Both the RDTI and RDLTC schemes require expenditure. Costs sitting on the balance sheet as an asset aren't expenditure — they're capitalised outflows. They don't count.

So this company had been doing genuine, qualifying R&D for two years, and it was entirely invisible to IRD.

The Path We Took

The RDTI General Approval for FY2026 wasn't available — the application window had passed. That ruled out the 15% RDTI credit for the current year.

But the RDLTC — the R&D Loss Tax Credit — was still on the table for FY2025, provided we moved quickly. The filing deadline was 30 April 2026.

We got the accountant on a call. To their credit, they engaged immediately. Once we walked through the IAS 38 position for a company at this stage, they agreed the capitalisation wasn't supportable. FY2025 accounts were refiled with IRD. The software development costs came back onto the P&L as an expense. The tax loss deepened — which is exactly what you want when claiming the RDLTC.

The claim was submitted with a week to spare.

The Roadmap From Here

Now we're working on the FY2026 RDLTC and, in parallel, building the General Approval application for RDTI — positioning FY2026 as Year 1 of what looks like a three-year R&D programme. That sets the company up for the full 15% RDTI credit on eligible spend as the programme matures.

Combined with the RDLTC years, the working estimate holds: approximately $1 million in credits over three to four years, from a 17-minute call.

If You're Capitalising Software Development Costs

Ask yourself three questions:

1. Can you genuinely assert a reliable useful life for this software right now?
2. Is your company's commercial viability — as IAS 38 requires — genuinely beyond doubt at this stage?
3. Have you ever applied for RDTI General Approval?

If the answer to the first two is uncertain and the third is no, you may be in the same position this client was in.

The RDLTC window for FY2025 is now closed. The FY2026 deadline is 30 April 2027. There's time — but the refiling and structuring work takes longer than most people expect.