Case studiesAI Automation · Voice Agent

Restaurant Reservation Voice Agent

A voice agent that answers a restaurant’s incoming calls, books, moves and cancels reservations, and writes every booking into Google Calendar and a customer database without anyone typing it in.

Talk to us about a system like this
  • 0%Calls handled
  • 0.0%Complete customer records
  • 0%Less admin time
  • 0Double bookings
Industry
Restaurants
Solution
Inbound voice reservation agent
Services
Voice booking · Availability check · Calendar and database sync
Stack
Retell AI · GPT-5 · Google Calendar API · n8n · Supabase
01The story
  1. Before

    Restaurants lose a good part of the day to the phone.

  2. Double bookings, customer details in three places, and a calendar kept by hand.

  3. After

    An agent now answers every reservation call.

One call, end to end: the ring, the conversation, and the two records it writes.
From a book full of crossings out to a confirmed table, in one call.
Voice AI AgentRestaurant ReservationsRetell AIn8n AutomationGoogle Calendar SyncSupabaseInbound Call HandlingBooking Automation
02How it works

One call in. Two systems updated. Zero typing.

The agent follows the same ten steps on every call, which is what keeps the data clean at the end of it.

  1. 01 · Caller
    Inbound call

    Someone rings to book, move or cancel a table.

  2. 02 · Retell AI

    The agent holds the conversation, with GPT-5 behind it.

  3. 03 · n8n
    WebhookCreate userCalendarSupabase

    Webhooks create the customer and the booking.

  4. 04A · Google Calendar
    Table for 4 · 7:00 PM
    04B · Supabase
    customer + booking saved
  5. 05 · Confirmation
    Table confirmed

    The caller is confirmed without anyone sending anything.

1Answer

  1. 01

    The call comes in: The agent answers and greets the caller.

  2. 02

    Party size: It asks how many people the table is for.

  3. 03

    Date and time: The caller gives the day and the time they want.

2Check

  1. 04

    Availability check: The agent checks the calendar for that slot before going any further.

  2. 05

    Special requests: Dietary needs, celebrations and seating preferences are captured here.

3Create

  1. 06

    Caller details: A first-time caller is asked for name, email and phone.

  2. 07

    Customer record: An n8n webhook creates the customer record.

  3. 08

    Booking: The reservation is booked against the exact timestamp.

4Sync

  1. 09

    Calendar: n8n creates the matching Google Calendar event.

  2. 10

    Database: n8n writes the booking metadata to Supabase.

03Sample call

One call. One booked table.

An example of the conversation, from the greeting to the confirmation.

  • Party size, date and time taken
  • Availability checked before the slot was offered
  • Special request saved with the booking
incoming callKateAI voice agent · example call
DeclineAccept
Tap Accept to hear the call.
04Challenge → Solution

What the phone was breaking.

Before

Staff answering the phone instead of working the floor

Every reservation meant stopping a task, taking details by hand, and writing them into a book.

After

The agent answers every inbound call and takes the booking from start to finish, so nobody has to leave the floor to do it.

The agent taking a reservation on the call, with book, reschedule and cancel actions.
Before

Two tables promised the same slot

A paper book and a busy service is how the same time gets given to two different people.

After

Before it confirms anything the agent checks live availability on the calendar, so a slot can only be sold once.

Google Calendar showing the requested slot as free before it is booked, and 0% double bookings.
Before

Customer details spread across three places

Names and numbers ended up in the booking book, in somebody’s phone, and in nobody’s memory.

After

Each caller becomes one record in Supabase, and the same record is recognised and updated the next time they ring.

The Supabase customers table, one row per caller, filled by the n8n webhook.
Before

The calendar kept by hand

Bookings had to be copied into the calendar after the call, which is where a fifth to a third of admin time went.

After

n8n creates the calendar event the moment the booking is confirmed, with the exact time on it.

The n8n flow from webhook to Google Calendar and Supabase, with the event it created.
Before

Special requests going missing

Allergies, birthdays and seating preferences were written in the margin of a page, when they were written at all.

After

The agent asks for them during the call and saves them with the booking, so the kitchen sees them.

A customer record captured by the agent, including the special request.
05Inside the system

The receipts.

Every slot is checked in Google Calendar before it is offered.One booking, written to the calendar and the customer record at once.The ten steps of a call, grouped the way the call moves.Bookings moved and cancelled on the call, with nobody on the phone.
Every slot is checked in Google Calendar before it is offered.
06Built with

Eight tools. One agent.

  • Retell AIVoice
  • 11Labs (Cimo voice)Voice
  • GPT-5 for dialogue logicConversation
  • n8n webhooksAutomation
  • Availability lookup workflowAutomation
  • Google Calendar APICalendar
  • SupabaseData
  • GPT-4o-mini for post-call analysisConversation
07What actually changed

The phone stopped being a job.

Every booking lands in the calendar and the database on its own.

The front of house got its attention back.

Synced after every call
Google Calendar0%
Supabase0%
Customer records0.0%
  • 0%
    Calls handledEvery inbound reservation call is taken by the agent.
  • 0.0%
    Complete customer recordsName, email, phone and any special request captured on the call.
  • 0%
    Less admin timeBookings no longer have to be written up and copied into a calendar.
  • 0
    Double bookingsAvailability is checked against the live calendar before a slot is confirmed.
  • 0%
    Bookings syncedCalendar event and database record created on every confirmed booking.
08Services

Voice agents that answer the phone.

Built on Retell AI and n8n, wired into the calendar and the database a business already uses.

Industries
Restaurants
  • Answers inbound calls and greets the caller
  • Takes party size, date and time
  • Checks live availability before confirming anything
  • Books, reschedules and cancels reservations
  • Captures dietary needs, celebrations and seating requests
  • Collects name, email and phone from first-time callers
  • Creates the Google Calendar event with the exact timestamp
  • Writes the customer record and booking metadata to Supabase
  • Sends the caller a confirmation message
  • Runs a post-call analysis pass over the transcript
Free 30-min consultation

Want one answering your line?

Tell us what your phone line has to handle and we will tell you, on a call, whether an agent is the right answer for it.

Talk to us about a system like this Chat on WhatsApp
  • 30 minutes on Google Meet
  • Usually reply in 2–4 hours
  • Talk to a senior engineer

More case studies

Pet Care Appointment Booking AgentAI Automation

Pet Care Appointment Booking Agent

A voice agent for grooming, vet and boarding appointments that books by phone, picks the right staff member and slot, and creates the customer in Square CRM without anyone re-typing it.

Outbound Database Reactivation AgentAI Automation

Outbound Database Reactivation Agent

An outbound voice agent that called more than 500 dormant event leads, asked what each business was struggling with, and sent the interested ones to a product demo.

Automated Content Generation PipelineAI Automation

Automated Content Generation Pipeline

A pipeline that watches a Google Sheet for new clients and produces their whole content pack, blogs, website copy, emails, ad copy and social posts, formatted and filed in Google Docs.