By now, you’ve seen what SR&ED eligibility looks like across industries — from AI and software to manufacturing, construction, and life sciences. This self-assessment pulls the eligibility test into one practical scorecard you can run against your own projects in a few minutes.
Score Each Statement (0 = No, 1 = Somewhat, 2 = Yes)
- We couldn’t predict the outcome of this project in advance, even with our team’s expertise
- We looked for an existing solution, method, or product and didn’t find one that fit
- Our team tested, measured, and adjusted their approach based on results, rather than following a fixed plan
- Something about this project didn’t work as expected at some point
- We generated new technical understanding, even if the end result wasn’t commercially successful
- This work went beyond styling, configuration, or minor customization of something that already existed
- Specialized technical staff (engineers, developers, scientists) did hands-on work on this project
- We have some form of contemporaneous record — notes, tickets, commits, lab logs — from while the work happened
How to Read Your Score
12–16: Strong candidate
This project likely involves genuine technological uncertainty and systematic investigation. A formal eligibility review is very likely worth your time.
7–11: Possible candidate
Parts of this project may be eligible even if the whole thing isn’t. A review can help isolate the eligible portion and quantify what it’s worth claiming.
0–6: Unlikely on its own
This specific project may not meet the bar independently — but it’s worth checking whether a related, more experimental phase of the work exists elsewhere in your operations.
This Scorecard Is a Starting Point, Not a Verdict
Self-assessment is useful for triage, not for a final answer — eligibility ultimately comes down to the specific technical details of the work and how well they’re documented. Many businesses score their projects lower than they should simply because they don’t yet see their own work the way a CRA reviewer does.





