Booking accuracy

Booking accuracy is whether an appointment an AI agent booked matches what the customer agreed to (the service, the day and time, and where the conversation settled them, the provider, location, party size and quoted price) and stood without staff having to fix it. None of the vendor pages read for Attify's research defines it; Attify's bookings standard measures it in two checks.

The two checks#

CheckWhat it asksEvidence
(c) Staff fixed itDid staff change the service, day or time, provider, location, party size or price within 48 hours of the booking (or before the visit, if sooner), or cancel and rebook it? Was the conversation handed to a person before the booking?The booking system's record of who made each change
(f) Wrong bookingDoes the appointment, as the business's system holds it, match what the customer agreed to in the conversation?An LLM judge, with a prompt published in the standard, on a random sample of 200 per business

A change the customer made through the platform's own self-service link is the customer's, not staff's. Where the system does not record who made a change, check (c) is not testable for that business, and the result says so.

Double bookings#

Under the vendor's own count, a second appointment for the same customer and service on the same day fails check (b), "not a new booking". In a reconstructed claim it is left out of the claim to begin with.

Where to read it#

Systems that record who created and changed each appointment make accuracy measurable: Calendly (scheduling method, canceler type), Cal.com (who cancelled or rescheduled), and practice-management systems such as Open Dental, which the standard lists as natural next connectors. Plain calendars (Google, Microsoft 365) do not record who changed an event.

Questions

How do I know if my AI receptionist double books?

Look for a second appointment for the same customer and service on the same day, and for appointments staff moved or changed within 48 hours of the AI booking them. Both show in a booking system that records who made each change.

Does a voice platform's 'call successful' mean the booking was right?

No. On Retell and Vapi, that flag is an LLM grading the transcript against criteria whoever built the agent writes. It is a self-report, not a booking record, and Attify's bookings standard never treats it as a claim or a check.

Sources

  1. Attify. Attify Audited Booking Standard v0.6: checks (c) and (f). Published 9 Nov 2026. Read 9 Nov 2026.
  2. Retell AI. Post call analysis. Read 31 Oct 2026.
  3. Vapi. Call analysis. Read 31 Oct 2026.