01
Give reception one answer to “what needs attention?”
Follow-up often disappears into message history, memory and disconnected notes. The product turns that work into a small queue with explicit states so reception can see what is due, what is waiting and what needs a person.
The operating model stays deliberately narrow. Alcora coordinates follow-up; it does not present itself as the clinic’s medical record or invent clinical truth.
02
Fit the channel before adding intelligence.
The workflow is WhatsApp-first because that is where much of the real follow-up already happens. Routine reminders, reschedules and recalls can be prepared from a known state without forcing staff into an unfamiliar communication system.
Sensitive language, ambiguous replies and anything that could change care remain visible for human handling. The product helps staff move the work; it does not make clinical decisions.
03
A small state model makes automation safer.
Four clear queue states are easier to reason about than a general assistant reading an unlimited conversation history. Each state can have an owner, an allowed next action and an exception path.
That structure also makes a pilot testable. The team can inspect whether the queue reflects reality before relying on reminders or adding more automation.
04
Separate modeled recovery from measured recovery.
The opportunity model uses appointments per day, working days, missed-appointment rate, an assumed recovery rate and the value of an appointment. It is useful for deciding whether a pilot deserves attention.
It is not a result. The product has not yet produced production evidence for recovered appointments, staff time, revenue or patient outcomes.