Skip to content
Docs home

OPERATIONS · UPDATED AUGUST 2026

Reminders

Three reminder waves — booking_created, day_before, two_hours — orchestrated with rate protection, retries, and a per-message status model.

The three waves

Every new booking plans three reminder waves at creation time: booking_created (immediate), day_before (the day preceding the appointment), and two_hours (two hours before). The wave set and its default offsets are defined in the reminder orchestration contract and enforced at the schema level — a wave outside the contract cannot be saved.

WaveWhen it firesPurpose
booking_createdAt booking creationConfirm the booking and share the management link
day_beforeDay before the appointmentReduce no-shows with a timely prompt
two_hoursTwo hours before the appointmentFinal prompt while the plan is still actionable

Recipients and channels

Reminder recipients can be the customer, the appointee, or an employee. Channels include SMS, WhatsApp, email, voice, and in-app notification, with the tenant's enabled channels controlling what actually delivers. Customers who have opted out of SMS reminders are skipped for SMS waves (the booking_created wave is exempt, since it carries the confirmation itself).

Rate protection and retries

The dispatcher enforces a per-minute SMS cap and retries failed sends up to a declared attempt limit before marking a message failed. Every planned reminder carries a status — pending, sent, failed, suppressed, or cancelled — so operators can see exactly what happened to each message rather than guessing.

Reminder ≠ delivery guarantee

QUE schedules, dispatches, and retries; carriers deliver. A message marked sent has left QUE. A message marked failed has exhausted its retries and is logged for operator review. Do not treat reminder success as proof the customer saw anything.

STILL READING

Something here still feels worth checking together.

This documentation describes the product exactly as it is implemented. If a behavior you expected is not here — or is here and does not match your deployment — tell the team. Implementation questions start on the contact page.