How to build a WhatsApp chatbot for your hotel

Start from what customers actually send. Guests ask about availability and rates for their dates, early check-in, airport transfers, directions, and what is included in the room.

Then look at what goes wrong today:

• OTA commissions eat 15-25% of every booking.
• Guests rebook through an OTA, not with you. The guest loved the stay. Next trip they book the same room through an OTA because you never gave them a direct way back.
• Return guests book via OTA again.

Each of these is a conversation with a predictable shape — a question, a lookup, a confirmation. That shape is what a chatbot handles well.

1. Dates — check-in and check-out.
2. Guests — adults and children, since room types and rates depend on it.
3. Room type — a list message of what is available for those dates, with the rate and what is included.
4. Add-ons — breakfast, airport transfer, early check-in.
5. Confirm and pay a deposit by link, or hand to reservations for anything unusual.

If the dates are full, the bot offers the nearest available dates or a sister property, rather than a dead end.

Notice what the business side does: it answers with the specific detail (time, reference, next step), it asks for one thing at a time, and it gives the customer a one-word way to reply. Those three habits matter more than how clever the bot is.

• Check-in and check-out times, early check-in and late checkout rules.
• Parking, airport distance and transfer prices.
• Breakfast hours and what it includes.
• Pool, gym and spa hours.
• Pet policy, extra beds, children's policy.
• ID requirements at check-in.

These answers never change mid-day, and every one the bot answers is one less call to the desk.

Reputoo supports both, and most hotels end up using both:

• Keyword flows — a fixed menu or trigger word ("MENU", "BOOK", "STATUS") leads down a set path. Predictable, easy to audit, right for anything with a fixed answer.
• AI chatbot — reads the customer's own words and answers from your knowledge base (website, price list, policies). Right for the long tail of questions nobody wrote a menu item for.

A sensible start for a hotel: keyword flows for the top two or three tasks, AI for everything else, and a person behind both.

A guest in the hotel messages from their own phone; the bot asks for the room number once and links the conversation to the stay. From then on, "extra pillows", "iron please" or "leaking tap" become tickets for housekeeping or maintenance, each acknowledged with "on its way, about 15 minutes" and closed when done. The bot should confirm the request is completed — or ask if it was — so nothing is silently forgotten.

The in-room menu can be sent as a PDF or catalogue; orders are taken with the room number, confirmed with the total and time, and sent to the kitchen. Restaurant table bookings for in-house guests work the same way. Charges go to the room.

• Complaints, refunds and anything where the customer is upset.
• Anything involving money the bot cannot see — a disputed bill, a custom quote.
• Health, safety or legal questions.
• Any customer who asks for a human. Do not make them ask twice.

For hotels specifically: rate negotiation for groups, complaints during the stay, and anything medical or security-related go to the front desk manager.

Handover on Reputoo moves the chat into the shared inbox with the conversation so far, so staff never ask the customer to repeat themselves.

1. Connect your WhatsApp number to Reputoo (official API; takes minutes).
2. Add your website and documents to the knowledge base so the AI answers in your hotel's own terms.
3. Build keyword flows for the top tasks above. What to connect: A property management system and a channel manager; the PMS reservation is what should trigger confirmations and pre-arrival messages.
4. Choose who gets handed-over chats, and during which hours.
5. Test from a personal phone, then share the number or a QR code with customers.

Guests often ask for late checkout or an extra night on the morning of departure. The bot can check the rules and availability, offer the options with the price, and confirm — or hand to the front desk if the hotel is full. It turns a front-desk queue at 10 am into a two-tap chat.

"The room next door is noisy" or "the room wasn't cleaned" need a person, quickly. The bot apologises, takes the room number and details, and hands to the duty manager with urgency marked. It never argues or offers compensation on its own.

Enquiries for ten rooms or a wedding go to the sales team with dates, room count, budget and occasion collected. The guest is told when the team will reply. The bot should never quote group rates.

Hotels receive messages in many languages. Keyword flows handle common words in the main ones; the AI chatbot understands the rest and replies in the guest's language. Room and facility names should be consistent across languages so requests are routed correctly.

• Book for different dates, guests and room types, including sold-out dates.
• Send in-stay requests as a guest and check each reaches the right department.
• Order room service and check the kitchen sees it with the room number.
• Ask for late checkout on a full day and an empty one.
• Complain and check the duty manager is alerted.

Watch direct bookings made in chat, in-stay requests closed within their target time, and calls to the front desk. Read handovers weekly; recurring questions the bot cannot answer are missing facts, and slow departments show up as requests left open.

Transfers are a frequent question and a small source of revenue. The bot quotes the price, takes the flight number and arrival time, confirms the booking, and sends the driver's details on the day. For guests arriving on their own, it sends a map link and directions from the airport or station, and landmarks for the last stretch.

Guests landing at dawn ask whether they can check in early. The bot explains the rule — guaranteed early check-in at a charge, or subject to availability — books it if chosen, and offers luggage storage and breakfast while the room is prepared. Setting expectations the night before prevents a tired guest arguing at the desk at 7 am.

A thank-you message after checkout, with the invoice and one question — "How was your stay?" with [Great] [Could be better] — catches problems while the guest still cares. "Could be better" goes to the front office manager. The same message can ask whether the guest would like to hear about future packages, which is the cleanest way to collect marketing consent.

Every handover arrives with the reservation or room number, the guest's words and the reason — a rate question, a complaint, a request the bot could not route. The bot tells the guest who will reply and when. Night handovers go to the night manager, not to a queue that waits until morning.

For direct bookings the bot sends a deposit link with the amount and the cancellation terms in one line, holds the room as provisional until payment, and confirms when it arrives. At checkout it sends the bill and a payment link so the guest can settle from the room. Refund requests and disputes go to the front office with the folio attached.

Before a long weekend or a local festival, update availability, minimum stays and any special rates in one place. The bot then quotes correctly, explains minimum-stay rules, and offers alternatives when the hotel is full. A bot quoting last month's rate on the busiest weekend of the year is how direct-booking trust is lost.

Accessible rooms, cots, dietary needs, a quiet room away from the lift — the bot records these on the reservation and confirms them back, so every team sees them before arrival. Requests it cannot guarantee are marked "on request" and passed to reservations, rather than promised.

Some questions decide whether a guest books at all: is my dog allowed, is there parking for an SUV, is the pool open in winter, is there a lift. The bot should answer these precisely and early, because a guest who has to wait for an answer books elsewhere. Keep them in the knowledge base with exact details — size limits, charges, seasonal hours.