Guest house · Samarkand

Jahongir Guest House — reservation and staff operations system

Bookings from every channel land in one operations flow — and the people who need to act find out in Telegram, not by being told twice.

Live in production

Guest house in SamarkandOTA and direct channels in one placeStaff workflows in TelegramFounder-operated production system

Business context

Jahongir Guest House sells through online travel agencies and directly. Reservation data lived in the channel manager, while the actual work — arrivals, special requests, breakfast counts, payments — lived in chats, on paper and in people's heads. The system connects the two.

Before

OTA bookingChannel managerSomeone noticesVerbal instructionsPaper notesMorning questions

Every change of plan had to be retold to each person it affected. On a busy day, someone was always the last to know.

The connected system

OTAs / directChannel managerWebhookOperations recordStaff TelegramKitchen countsPayments & receiptsReporting

Human approval points are marked in the automation table below — the system prepares, people decide.

Inside the system

Screenshots are being prepared with demo data and will be published after approval.

Expense tracking — every purchase with category, currency and approval status

Staff Telegram — the team learns about arrivals and changes automatically

Housekeeping Telegram — a checkout automatically notifies cleaning about the exact room

Kitchen bot — tomorrow's meal counts reported in Telegram

Daily cashier audit — the system itself flags forgotten shifts and stale FX rates

AI guest assistant — prepares booking actions but executes nothing until the guest confirms

What runs automatically — and where people stay in control

Trigger	System action	Human checkpoint	Fallback
New OTA or direct booking	Creates or updates the operations record	Staff check special requests	Manual entry
Booking change or cancellation	Updates the same record; failed syncs are flagged	Operator reviews flagged syncs	Manual correction
Online payment confirmed	Sends the receipt and posts the payment to the channel-manager invoice	Finance review where required	Manual reconciliation
Guest writes in Telegram or WhatsApp	AI assistant drafts a reply with real availability and prices	Operator supervises; the assistant cannot book on its own	Staff reply directly
Breakfast approaching	Kitchen bot reports guest and meal counts in Telegram	Cook confirms the numbers	Verbal count
Guest checks out	The cleaning channel gets a room-specific notification	Housekeeper plans the room turn	Verbal instruction
Every morning	A cashier audit reports to Telegram: open shifts, stale FX rates, sync status	Manager acts on the findings	Manual checks
Staff records an expense	Categorised with currency and approval status	Manager reviews the reports	Paper notes

What changed

One view of every booking regardless of sales channel
Staff learn about arrivals and changes without anyone retelling them
The kitchen knows tomorrow's numbers without a morning round-up
Payments and receipts stay reconciled with the channel manager
Sync failures surface for repair instead of hiding until check-in

What remains human

Pricing exceptions and special offers
Guest complaints and sensitive conversations
Refunds and disputed payments
Anything the AI assistant is not sure about
The final word on room moves

Privacy and reliability

Published screenshots use demo or anonymized data. The production system runs with role-based permissions, operational records, backups, failure monitoring and a manual fallback for every automated step.

Automation stories from this platform

Live in production

Payment link to automatic receipt

The guest gets confirmation without waiting for staff

See how it works →