"Scientific or technological uncertainty" sounds far removed from medicine. In practice, it describes a familiar situation: your team reaches a problem that the published evidence, standard methods, and its own existing knowledge cannot answer.
A plain-language test
Ask this question:
Could a qualified person, using what your team already knew and the public information reasonably available, predict whether the proposed approach would work?
If the answer was already available, carrying out the work may still have been difficult, but it was probably routine for SR&ED purposes. If the answer could only be found through a systematic experiment or analysis, the project may have faced the right kind of uncertainty.
The gap must exist in the scientific or technological knowledge available to the team and through reasonable public sources. One person's lack of experience is not enough. A trainee learning how a validated assay works is gaining existing knowledge. A team testing an assay because nobody knows how interference in a new sample type will affect it may be generating new knowledge.
What this can look like in medicine
A published protocol may need to be adapted for a paediatric population when adult data cannot predict the dosing behaviour. An imaging method may work on one scanner but fail on another for reasons that are not understood. A diagnostic assay may behave unpredictably in a sample type it has never been validated against.
These examples do not qualify simply because they are medical or complex. What matters is that available knowledge could not provide the answer and the team tried to find it through experiment or analysis.
That investigation should have a clear thread. The team starts with a possible explanation or solution, tests it, records what happened, and uses the result to decide what comes next. Trial and error can be part of that process when each attempt is planned and each result informs the next one.
The project does not have to succeed. Learning why a proposed solution fails can still add useful knowledge.
Make the unknown visible
A reviewer sees the work long after it happened. Your records should show what was unknown at the start, which approaches were tested, what the team observed, and what it learned.
Lab notebooks, protocol amendments, failed-run data, analysis files, and meeting notes often contain that story already. The important part is keeping them dated and specific enough to show how the thinking changed over time.
Grant applications tend to emphasize why a project is likely to succeed. An SR&ED description has a different job. It explains what the team genuinely could not know and how it tried to find the answer. A short, confidential fit conversation can help determine whether your records show that kind of uncertainty and whether the work is worth assessing further.