A booking bot is a Telegram bot that shows free slots, books the visit and reminds the client about it. It takes the endless when-can-I-come messages and manual rescheduling away from your front desk, but only on one condition: the bot reads the schedule from the same system your administrator works in, not from a separate spreadsheet.
Key takeaways
The bot covers the whole booking cycle: the client picks a service and a specialist, sees real free slots, confirms the booking and gets a reminder before the visit. The administrator steps in only for non-standard cases.
The same scenario works for clinics and dental practices, beauty salons and barbershops, car services, training centres and independent professionals. The rules differ: service duration, gaps between visits, room preparation, limits on who can perform the service. The skeleton of the bot stays the same, and that is where Telegram bot development for business starts.
A good booking takes five or six taps and one phone confirmation. The longer the chain, the more people abandon it halfway and call the front desk instead — or go where booking is simpler.
After that the bot runs without a human: it reminds the client about the visit, processes cancellations and asks for a rating afterwards. How bookings and clients reach the CRM and what happens to them next is covered in the article on Telegram bot and CRM integration.
The bot does not keep the schedule itself — it asks the system the administrator works in. That can be an industry booking platform, a medical information system, a CRM with a calendar or a shared calendar over CalDAV.
Such a link needs three things: the slot is held while the booking is being completed, the booking is created in a single operation — otherwise two clients take the same time — and changes made in the system flow back to the bot. If the administrator moves a visit, the client should learn about it from the chat, not from a phone call.
There are two ways to show the time grid. The simple one is buttons under a message: reliable and works on any device. The richer one is a Mini App, a web interface inside Telegram that fits a month-long calendar, specialist cards and filters. Mini Apps come with a mandatory rule: user data arrives in the initData parameter, and the server verifies its HMAC-SHA-256 signature with the bot token before creating anything.
Tip Agree in advance which system is the master one. The schedule is kept there, and the bot only reads free time and creates bookings through the API. Two parallel master spreadsheets produce double bookings within the first week.
A reminder before the visit is the simplest part of the bot and usually the one that pays off first: the client either shows up or frees the time in advance, and someone else takes the slot.
The working pattern is two reminders: one a day ahead, so the client can replan the day, and one two or three hours before, so they leave on time. Each reminder carries confirm, reschedule and cancel buttons; a tap changes the booking status in the schedule and the CRM instead of simply messaging the administrator.
Broadcasts have a limit worth remembering at launch: only people who started the bot themselves receive reminders. That is why the link goes into the administrator's signature, the website and company listings, while the first booking is often still made by phone.
Cancellation rules are set before launch, otherwise the bot becomes a way to block time for free. The minimum is a window in which cancelling is free and a clear text about what happens if the client does not show up.
Prepayment in Telegram is taken by card through a payment provider — in Uzbekistan these are Payme and CLICK. How the payment is connected and what has to happen right after the charge is covered in the article on payments in a Telegram bot.
For service bookings a Telegram bot beats a website form on the speed of repeat visits, and a mobile app on the fact that nothing has to be installed. An app makes sense when booking is only one of the product's functions.
| Criterion | Telegram bot | Website form | Mobile app |
|---|---|---|---|
| How booking starts | A link or a QR code, no install needed | A visit to the site, often from an ad | Install from an app store |
| Repeat booking | The chat is still there, two or three taps | Details are entered again | Fast, unless the app was deleted |
| Reminders | A chat message, no separate fee | SMS or email, paid separately | Push, if notifications are allowed |
| Cancelling and rescheduling | Buttons in the same chat | A link from an email or a call | A screen with the list of bookings |
| What you maintain | The bot server and the schedule exchange | The site and the form | Two app versions and store releases |
| When to choose it | Services with repeat visits | First touch and advertising | A product where booking is part of a larger service |
These channels do not compete: the website form catches the first touch, the bot takes repeat visits and reminders. The only thing that matters is that both write into one schedule.
In Uzbekistan Telegram is an everyday channel, so booking through a bot asks nothing new of the client: in October 2025, 89.0 % of the country’s population used the internet, according to the DataReportal Digital 2026 report. A local launch does have a few details worth planning for.
A booking bot usually fails to take root not because of technology, but because it was launched next to the working process instead of inside it.
Mistake The bot replies with a thank-you and a promise to call back — that is a request form, not a booking. The client still waits for the call, the administrator still dials, and the queue has not moved anywhere.
A bot that takes requests and bookings and creates them in the CRM starts at $1,200; the scope of work is described on the page for bots and AI assistants. The final figure depends on how many roles, services and rules the bot has and which system it exchanges data with.
Three things move the budget most: how complex the scheduling rules are, how many integrations there are and whether the bot needs payments. A bot for one service and one specialist is assembled quickly; a chain of locations with different price lists and prepayment is already a project.
We start with the scenario rather than the code: we walk the client path from the first message to the reminder and look at where it breaks today. Then we connect the bot to the system that already holds the schedule, create bookings in the CRM and set up reminders with confirmation buttons.
After launch we read the first conversations: where people stop, which questions still go to the administrator, which wordings have to be rewritten. The metrics are simple — the share of bookings made through the bot, the share of cancellations and no-shows, and the time from the first message to a confirmed booking.
Let’s discuss your project
Tell us what you need, and we will estimate the timeline and cost and suggest a solution.
Yes. At the start the bot can create bookings directly in the scheduling system and notify the administrator in a chat. But then client history lives only in the schedule: nobody counts repeat visits, booking sources or refusals. A CRM is usually connected as a second step, once the bot already produces a flow of bookings.
Buttons are enough for a single practice or salon with one or two services: choosing a specialist, a day and a time fits into a few messages. A Mini App is needed where the client has a wide choice: a chain of locations, dozens of services and a month-long calendar. You can start with buttons and add a Mini App later without rewriting the bot.
The administrator creates the booking manually in the same scheduling system, from the client card. The bot and the administrator work with one calendar, so such a booking takes the slot immediately and never conflicts with online bookings. Reminders for that client are sent by SMS or by phone.
No. The bot sends messages only to people who opened it and pressed the start button themselves. That is why the link is placed where the client already talks to the company: the administrator's signature, the website, social profiles and a QR code in the room. After the first booking the person stays reachable and reminders go out automatically.
The same administrator. The bot does not replace a person, it takes over the repetitive part: showing free time, booking, reminding and accepting cancellations. Complex cases stay with the administrator — moving a group of bookings, disputes about time, matching a specialist to a request.
Compare three numbers before and after launch: the share of bookings made without the administrator, the share of no-shows and the time from the client's first message to a confirmed booking. The last one usually moves first: bookings start arriving at night and at weekends, when nobody picks up the phone.
Sources
Cover photo: Jakson Martins, Pexels
