SR&ED
Rejected claims describe a product. Approved claims describe a problem nobody knew how to solve.
The single biggest predictor of a successful SR&ED claim is not how hard the engineering was. It is whether the write-up describes the work in the language the program is built around: uncertainty, hypothesis, systematic investigation, advancement.
Building something difficult is not, by itself, eligible. The question is whether a competent professional in the field, using existing knowledge and standard practice, could have known in advance whether the approach would work and how to achieve it. If yes, that is engineering. If no, that is technological uncertainty, and the work done to resolve it is what you claim.
Two descriptions of the same project
Not eligible as written: "We built a real-time recommendation engine for our marketplace." Eligible as written: "Existing collaborative filtering approaches degraded below acceptable latency at our sparsity level. It was not known whether the ranking quality could be preserved under that constraint. We formulated and tested three approaches and measured the tradeoff."
General information for Canadian founders, current to 2026. It is not tax advice and does not account for your specific facts. Rates, thresholds and rules change, confirm the current figures before acting.
A short call is usually enough to tell you whether there is anything here worth acting on.