Evidence-led executive learning · reviewed August 2026
Annotated opportunity decision
Fictional comparison using ranges and evidence notes rather than false precision.
| Candidate | Evidence summary | Gate | Why |
|---|---|---|---|
| Internal knowledge assistant | Moderate service value; bounded task; approved corpus exists but lacks named content owner; staff verify source-linked drafts; high reversibility. | Prepare then test | Resolve source ownership in two weeks, then test a representative question set and staff workflow. |
| Customer recommendations | Potentially high growth value; weak baseline; integration, consent and measurement uncertain. | Prepare | Instrument the current journey and resolve data/measurement assumptions before product testing. |
| Hiring screening | Capacity hypothesis; high affected-person consequence; weak representativeness evidence; significant contestability and assurance needs. | Do not pursue now | Necessity, alternatives and protections have not been established. |
Sensitivity
If source ownership for the internal assistant cannot be established, its evidence readiness falls and the customer opportunity’s baseline action becomes the better first investment. The ranking is conditional, not permanent.
Weak counterexample
“Hiring screening scores 23/25 and wins.” No criterion evidence, confidence, consequence gate, dependency or sensitivity is visible.