There's a line in a recent SAAS NORTH interview with Chelsey Fawcett, CEO of Grant Solutions Canada, that lands harder than it first appears:

“Eighty percent of my job isn't grant writing.”

The rest, she says, happens well before anyone opens an application portal: getting a business clear on what it's trying to achieve, and on how an outsider would measure whether it did. Her argument is that grant readiness now looks a lot like investor readiness, and that the companies who keep winning funding are the ones who got organized long before they needed to be.

We'd add one thing from the accounting side.

A good share of that other 80% isn't strategy. It's record-keeping. And unlike a story, which you can write in a month, that's something your books either already do or they don't.

First, this isn't a tech story

It's easy to assume government R&D funding is for software companies. The numbers say otherwise.

In the CRA's most recent published year, software development accounted for 42.6% of all SR&ED credits awarded. Which means nearly 60% went somewhere else: medical and mechanical engineering, electrical work, materials, chemistry, environmental science, agriculture, food processing.

And these aren't big-company programs. 65% of claims came from businesses with under $4 million in gross income.

If you're improving a product, a recipe, a material, or a process and you're genuinely unsure at the outset whether your approach will work, you're likely in scope. A food producer reformulating to extend shelf life, a manufacturer trying to cut waste on a line, a fabricator testing a new joining technique: all of that is the same conversation as a software team building something novel.

All figures below are Canadian dollars.

Two deadlines, and both start before you apply

Canada's two main R&D funding programs work in opposite directions. The timing gap between them is where most money quietly disappears.

IRAP pays for work you haven't done yet. The National Research Council's program funds projects going forward. You work with an advisor, get approved, sign an agreement, then draw money as you hit milestones. The signing date is the line: money you spent before it generally doesn't count, and there's no going back for it. Start the work first and that spending is simply lost for IRAP purposes.

SR&ED pays you back for work you've already done. You do the work, keep records, and claim afterward through your corporate tax return. The deadline is firm - 18 months after your fiscal year-end, with no extensions.

But here's the catch on the second one. Your claim rests on records created while the work was happening. Notes on what you tried, what failed, who spent time on what. Written afterward from memory, it isn't evidence.

So one program needs you organized before you start. The other needs you to have been organized as you went. Neither can be rescued by a good writer at the end.

That's what “readiness starts earlier than you think” actually means in practice. It isn't a mindset. It's two deadlines pointing in opposite directions.

You can't be paid twice for the same dollar

This is the rule that produces the most unpleasant surprises, and it's simpler than it sounds.

Most owners assume grants and tax credits stack: take the grant, then claim the credit on the full project cost. They don't work that way. If the government already paid for something, you can't claim a credit on it as well. Accountants call the reduction “the grind.”

Here's what it looks like. Say a grant covers $200,000 of wages on a project. Those wages come out of what you can claim. At the top SR&ED rate of 35%, that's roughly $70,000 of credit you can no longer claim on those particular dollars.

You're still better off taking both. The programs are built to work together, and the combined result beats either on its own. But it isn't “grant plus full credit” and if that's how you built your cash forecast, your forecast is wrong by a real number.

Two related points worth knowing:

• Provincial R&D credits count as government help too, so federal and provincial don't simply add up either.

• There's more than one way to calculate overhead in a SR&ED claim, and which method you choose changes how the reduction applies. That's a genuine “ask your accountant before you file” question.

One thing that changes how much you get back

The 35% rate isn't automatic, and the difference matters more than most owners realize.

That rate applies to Canadian-controlled private corporations - broadly, private companies controlled by Canadian residents. It's also refundable, which is the important word: the CRA pays it out in cash even if you owe no tax.

Companies that don't qualify, including foreign-controlled ones, generally get 15%, and it's non-refundable — a credit against tax you might not owe for years.

For a business that isn't yet profitable, that's close to the difference between money and no money.

What tips a company out of the first category isn't always obvious. It comes down to who controls the business, and that can turn on more than share counts - options held by non-residents, or agreements giving outside investors a veto, can all factor in.

The practical takeaway: if you're taking on outside investment, ask what it does to your R&D credit before you sign. It's a quick question for your accountant and lawyer, and an expensive discovery at filing time.

What this asks of your records

To claim correctly, you have to answer one question: which exact dollars did the grant pay for?

If your books can't separate them, you end up somewhere bad. Claim too little and you leave money behind. Claim too much and you're exposed on review and the burden of proof is yours.

In plain terms, four things:

1. Costs tracked by project, not just by category. “Wages: $1.2M” tells you nothing here. You need to know who worked on which project, and for how much of their time.

2. Time recorded as it happens. Not reconstructed at year-end. The split between grant-funded and self-funded work has to rest on real records.

3. Notes kept during the work. What you were trying to solve, what you tested, what didn't work. The whole basis of a claim is that the outcome was uncertain and that's only provable with notes written while it still was.

4. Grant agreements filed with the accounting. The reduction is applied based on what the grant was for, so the wording of the agreement matters. Keep those documents where the numbers live.

Bringing it back to the strategy

Chelsey's starting point that funding begins with spotting friction in your business, and that these programs exist because governments want certain things to happen faster, is right. The storytelling is real work.

But stories and evidence have different lead times. You can rewrite a story in a month. You can't create a timesheet after the fact, and you can't make finished work eligible for a program that only funds what's ahead.

It's the same pattern we see on the tax side constantly: most of what looks like a filing problem at year-end was a bookkeeping decision months earlier.

None of it is hard. It's just very difficult to add later.

A quick readiness check

If you're thinking about government funding in the next year or two:

• Could you show labour costs by project for last year, without rebuilding them from scratch?

• Do you have notes written during the work, not after?

• If you took a grant, do you know which specific costs it's considered to have paid for?

• Do your books separate project work from general payroll?

• Do you know your SR&ED deadline as an actual date — 18 months after your fiscal year-end?

Every “no” on that list is fixable. It's just much cheaper to fix now than to discover during a review.

This article is general information, not tax or accounting advice. Eligibility and program interactions depend on your specific circumstances, the choices you make when filing, and how your funding agreements are written. Talk to a qualified advisor about your situation.

Planning a grant application or an R&D claim this year? Get in touch, the bookkeeping decisions that determine the outcome are the ones you're making right now.

Sources

• SAAS NORTH NOW, “Why Funding Readiness Starts Earlier Than Most Founders Think,” interview with Chelsey Fawcett, CEO, Grant Solutions Canada (Kat de Sousa)

• Canada Revenue Agency, SR&ED Tax Incentives — Annual Program Statistics (April 1, 2025 – March 31, 2026)

• Canada Revenue Agency, SR&ED Tax Incentive Program — Form T661 filing requirements and deadlines

• Income Tax Act, subsections 127(11.1) and 127(18) — government assistance and the reduction of qualified expenditures

• EY Canada, “IRAP or SR&ED? That's the question.”

• National Research Council Canada — Industrial Research Assistance Program (IRAP)

• Bill C-15 (Royal Assent March 26, 2026) — SR&ED expenditure limit and eligibility amendments