Review questions
Integrated summary: Diagnosing adoption opportunity and work context
We turned a demand to “use AI” into a defined job, outcome, guardrail, and non-AI alternative. The workflow map locates the suggestion, source of truth, and human authority; it also exposes waiting time that faster drafting cannot remove. A baseline fixes start and end points, denominators, and case mix. An overall mean can rise while each group improves when the mix changes. Stakeholder diagnosis separates skill gaps, incentives, access problems, and justified objections, testing competing explanations instead of labeling people resistant. For pilot selection, data permission, authority, and harm control are eligibility conditions before benefit ranking. Scores then need sensitivity checks and comparison with non-AI repair. The resulting decision names a bounded first use case, evidence that would reverse the choice, and a stop owner. Success in that use case does not transfer automatically to recruitment or customer decisions.
Field observation expands the picture to work missing from system records, including expert help and pre-ticket corrections. Separate effort from waiting and test a competing explanation before choosing a tool. Between branches, compare input, purpose, authority and case mix; test changed components while preserving sound evidence for unchanged parts. Commitment decisions weigh learning against waiting and reversal costs, treating past expenditure as accounting and learning information. Adoption also requires effective access to sources, help and escalation. Together these links support a bounded trial whose result can change the organizational decision.
Review questions
Test your understanding with these questions. Marking and saved results are available with a free account.
Create a free account to mark your answers and save your progress.