Promise: decide today whether a controlled £7,500 pilot is worth running—not whether reminders should roll out. Narrative: measurable problem → bounded learning opportunity → safeguards → decision.
1 (1:00) “May's no-show baseline was 214 of 1,840 appointments”—show count and 11.6%, plus Finance-method £18,900 estimate; purpose: size the observed problem without claiming causality. 2 (1:30) “A four-week, two-clinic pilot can test the unanswered question”—diagram scope and comparison plan; state vendor effectiveness is unknown. 3 (2:00) “Privacy approval is a launch gate, not an assumption”—show date/time-only content, review owner missing, and no-send condition before approval. 4 (2:00) “Success needs clinical, operational, and cost measures”—define delivery, opt-out/complaint, no-show comparison, workflow burden, and data-quality checks; thresholds need approval. 5 (1:30) “The pilot costs £7,500 and does not authorize rollout”—quote, included scope, missing commercial details. 6 (1:30) “Two clinics are willing; one operational owner is still needed”—roles and dependencies. 7 (1:30) “Decision: budget, owner, privacy gate, and measurement criteria”—four explicit approvals. Reserve 1 minute contingency inside the 12.
50% cut: slides 1, 2, 3, 7. 75%: add 4 and 5. Q&A prepares cost method, message privacy, comparison validity, and stop rules.