Blueprint an AI-Assisted Clinic Booking Service
Build an evidence-backed current-state blueprint, locate failure mechanisms across the visibility line and design a testable recovery-aware future state.
27 minutes
Scenario
A clinic network introduced an AI booking assistant. It offers appointments quickly, but referral validation, accessibility requests and rescheduling context fail across hidden hand-offs.
- Your role
- Service designer preparing a controlled pilot
- Method
- Service Blueprint
Evidence pack
Customer action
Patients upload a referral and accessibility needs before selecting a slot.
Frontstage
The assistant confirms a slot before referral validation finishes.
Backstage
Coordinators manually reconcile failed referral matches twice daily.
Integration
Accessibility notes are absent in 12% of clinic-system records created through the assistant.
Recovery
Rescheduling creates a new case and discards the original exception history.
Outcome
Median booking interaction is four minutes, but 9% of bookings cannot be used without staff rework.
Constraints
- Protect health and accessibility data.
- Preserve a non-AI contact path.
- Do not remove clinical or eligibility controls.
Case steps
Work through each prompt using the evidence pack. These guided cases support self-directed practice; server-scored attempts are not available yet.
Define the patient segment, scenario, trigger, usable completion and exclusions.
Map customer/evidence, frontstage, backstage and support layers for the current state.
Identify three failure points and explain their customer and operational effects.
Design and test a future state with recovery, owners, measures and human control.