Pick the unknown.
Then pick the method.
A technical exercise should be chosen for the decision it needs to support.
If the plan depends on a specific capability, begin with a feasibility trial. If the choice is between plausible approaches, use an option study. If the uncertainty sits along a working path, an evidence prototype can make that path observable.
These routes can overlap, but they should not blur the purpose of the engagement. A clear first question helps determine which implementation and evidence belong in scope.
Can the required part actually work?
A bounded technical exercise around the uncertainty that could change the plan.
Which option fits the actual conditions?
Compare plausible approaches using the same task, constraints, and evidence frame.
What does a working slice need to establish?
Build the smallest technical slice that can expose a consequential assumption.
Check the conditions
for a useful answer.
Confirm that the relevant access, permitted samples, platform capabilities, and decision owner are available. A trial built around a substitute may be appropriate, but its boundary needs to be explained before the result is interpreted.
The deliverable can include an implementation artifact, but it should also explain the method and the limits of the finding. Prototype reuse, production implementation, and ongoing support are separate scope decisions.
Plan the investigation