A voice agent answers the salon's calls day and night and books clients into YClients or another booking system
A typical salon chain loses up to 40% of its calls in the evenings and at weekends. The agent picks up at any hour, finds a specialist and a free slot, and books the client into the software the salon uses to manage bookings, such as YClients. If a question is tricky, the agent passes it to a receptionist.
A client calls at 21:30, and nobody picks up
Clients of a premium salon call to book whenever it crosses their mind: on a lunch break, on the way home, before bed. If nobody answers, they message another salon on WhatsApp and often never come back.
Receptionists usually take calls from 10:00 to 21:00. After that, calls go to voicemail, and nobody listens to it until the morning. By the time the salon calls back, the client has often booked somewhere else.
The first thing salons usually do is hire an extra receptionist for evenings and weekends. At market rates, that costs about 95,000 ₽ a month per salon. A new person needs time to learn the specialists, rooms and treatment times. And while they are talking to one client, another client's call still goes unanswered.
The agent talks to the client and books them into YClients itself
Clients keep calling the salon's usual number, and the calls are forwarded to the agent. Here is what happens during a call:
Vapi or Bland AI, a voice agent platform, answers the call. If local law requires consent to record the call, the agent asks for it first
GPT-4o or Claude knows 14 typical scenarios: a new booking, moving or cancelling a visit, a question about prices, choosing a specialist and more
Passes the agent's commands to YClients: find a free slot, make a booking, open the client's record and visit history
The client is booked into the right salon and room. The connector checks that the slot is free
The client gets an SMS confirmation. If the agent isn't sure it has understood the client, the call goes to a receptionist
What a client can ask for
A client might say: "I'd like Katya for a manicure, but not before Thursday evening", "can I move my Tuesday appointment to Friday", "how much is coloring with a top stylist", "is there a slot for a facial massage this week".
For each salon in the chain, I write separate instructions for the AI model. At the start of each call, the agent loads the services, specialists, treatment times, rooms and prices from YClients.
When the call goes to a person
In a premium salon, the phone call is part of the service too. So if the agent still hasn't understood the client after two follow-up questions, it transfers the call to a receptionist during the day or adds the client to the callback list at night. It does the same when it isn't sure of its answer. During the first two weeks after launch, I use call recordings to adjust how strict this check is.
How the agent works with YClients
The connector service gives the agent a set of commands. The agent decides which one to use and in what order, and the connector checks each command before passing it on to YClients.
YClients is the example here. Another booking system with an API connects the same way. If your salon uses Fresha, Booksy or other software, I'll check how to connect it during the audit.
SMS confirmation
After booking, the client gets an SMS with a link to their booking in YClients. From there they can check the time, reschedule or cancel. In YClients the booking looks exactly as if a receptionist had made it.
What the agent is built from and where the data is stored
Voice agent platform: answers the call and brings speech recognition, the AI model and the agent's voice together in one conversation
AI model: runs the conversation and picks the right commands
Speech synthesis service: reads the agent's replies aloud in a natural voice
The agent uses it to book, move and cancel visits directly in the salons' schedule
Connector service: checks the agent's commands, handles access to the booking system and sends each booking to the right salon in the chain
Sends the client a text with a link to the booking. I connect a provider that delivers in your country
Runs the connector service and holds the database. I work with any VPS
Database of call transcripts and quality scores. I use them to improve the agent's instructions
Clients' names and phone numbers are stored on the server with the connector service. I place this server where your country's data protection law requires, for example in the EU.
What to expect: an estimate for a chain of three salons
All figures below are an estimate based on typical numbers for small and mid-sized businesses in Russia, so they are in rubles. For your salon, I'll recalculate them from your call statistics, average bill and currency.
today → target; all calls, including those between 21:00 and 10:00 and at weekends
a month, if the agent brings in an average of 41 bookings outside working hours
a month per salon. An evening and weekend receptionist costs about 95k at the market rate. The agent costs about 8k: minutes on the voice platform, the ElevenLabs voice, AI model requests and the server. Setup and support are paid separately
The key figure in the estimate is 41 new bookings a month for each salon. These are clients who call in the evenings and at weekends. Today their calls go unanswered.
The daytime receptionist carries on working at the front desk and greeting guests.
How long the launch takes
from the start of work to launch. By my estimate, the agent pays for itself in 2–3 months through bookings that are lost today
off the price list and a 90-day warranty. In return, I ask for permission to describe the project without naming the company
for 5,000 ₽: I'll count how many calls the salon loses today and recalculate the figures for your salon
Prices for voice agents and online booking are on the Pricing page.
Where else an agent like this helps
The building blocks are the same everywhere: a voice agent, an AI model, booking software and a hand-over to a person. For another business, only the connection to its booking software or CRM changes:
- → Barbershops, nail studios, massage rooms — if bookings are kept in YClients, Altegio, Beauty Pro or another system with an API
- → Fitness clubs and studios — booking and cancelling classes, moving memberships
- → Auto repair shops — booking into Alpha-Auto or the shop's own CRM
- → Language schools, tutors, children's centers — booking a trial lesson
- → Delivery and cleaning services — taking and rescheduling orders outside working hours
Blog article on this topic: Booking chatbot: where it breaks and nine questions for the vendor
Related cases and ready-made solutions
A voice agent for dental clinics that spots acute pain and books patients into the schedule
A patient calls or writes on WhatsApp, and the agent asks eight questions to work out how soon they need a…
Reviews on Yandex Maps and 2GIS: the shop asks every customer for a review, and AI helps draft the replies
Eighteen hours after a work order is closed in Alpha-Auto (1C), every customer gets the same WhatsApp…
After a missed call, the customer gets an SMS with the real ship date
The phone system reports a missed call, and the service finds the customer and their orders in two Shopify…
Start with the 5,000 ₽ audit: a written report and a payback estimate
I'll tell you what to automate first, when it will pay for itself and whether your task needs AI at all. Sometimes it doesn't.
Or just ask a question. During business hours I reply within 2 hours