AI Concierge, Chatbot or Guest App, What Your Hotel Actually Needs
November 9, 2026 · 4 min read · Mateo Bolívar
If you are weighing an AI concierge, chatbot or guest app, start with the tasks that take your team's time today, not with the name of the tool. Each one solves something different, and what your hotel often needs is a combination with clear limits.
In this article I compare the three by task, by channel and by what they need to work. At the end you have a table to fill in with your team.
What each one is, without caricatures
A chatbot is a program that converses in text. Some are very simple and follow a menu of options, and some understand freely written questions. A good chatbot can answer your hotel's frequent questions well. Its limit depends on how it is built and on the information it has, not on the word chatbot.
A guest app is where the guest sees and requests things from the hotel, such as the menu, room service, a spa booking or a housekeeping request. At Contler this is called Hotel Apps, and it works as a web app, with nothing to download.
An AI concierge handles guest requests in conversation. In Contler's case, the AI Concierge answers by chat or voice 24/7, with the hotel's brand, and the guest gets in with a QR code, NFC or a link. Staff receive cases in their own app, on iOS and Android, and the agent alerts the team when a person is needed.
AI concierge, chatbot or guest app, compared by task
The most useful way to decide is to look at what happens with each concrete task.
- Answering breakfast hours or the wifi password. A well-fed chatbot can do it and an AI concierge does it. The app can show it if the guest looks for it.
- Asking for an extra towel or pillow. The app lets the guest make the request if they find it in the menu. The AI concierge receives it in conversation and passes it to the right area.
- Seeing the menu and ordering room service. This is the natural ground of the guest app, because the guest wants to see photos and options.
- Answering in the guest's language. Contler's AI Concierge answers in more than 20 languages. Check this point in any tool you evaluate.
- Answering the phone. A text chatbot does not take calls. That is what a voice agent is for, which at Contler is Voice Reservations, with a different scope from the concierge.
- Solving a complaint or an emergency. None of the three should solve it alone. What matters is that the case reaches a person quickly with full context.
What each one needs to work
All three depend on the same base, the hotel's information. Hours, services, policies and conditions must be confirmed and have an owner. I wrote about how to organize it in this guide to preparing an AI concierge.
The differences show up in the rest. The app needs you to load menus and content. The concierge also needs clear rules about which cases go to a person. And any task that depends on a reservation, like changing dates, needs integration with your systems. I explained that boundary in the article on an AI concierge without a PMS.
When to combine them
In many hotels the answer is not to pick one. The guest sees the menu in the app and asks in conversation for what they cannot find. The team receives orders and requests in one place.
What you should avoid is mixing scopes. If the hotel announces an assistant that answers everything but only has a menu of options, the guest gets frustrated. If it announces an app and what exists is a conversation, the same happens. Name each thing for what it does.
A table to decide with your team
Copy this table into a sheet and fill it in with the front desk and each department head.
- Column 1. The task, written the way the guest would say it.
- Column 2. Roughly how many times a week it comes in.
- Column 3. Whether the answer is the same for everyone or depends on a reservation.
- Column 4. Whether the guest prefers to see options or to ask.
- Column 5. Whether it should ever go to a person.
Frequent tasks, with the same answer for everyone and that guests prefer to ask about, belong in conversation. Those the guest wants to see, like the menu, belong in the app. Those that depend on a reservation wait for integration. Sensitive ones always have a path to the team.
Illustrative example. A 40-room hotel fills in the table and finds that half of its night messages are about hours and services, and that almost all food orders come by phone. It decides to start with conversation for questions and the app for the menu, and leaves reservation changes with the front desk.
Frequently asked questions
Is a chatbot worse than an AI concierge?
Not necessarily. It depends on which tasks it solves and with what information. Compare by task, not by label.
Does the guest have to download anything?
With Contler's AI Concierge and Hotel Apps, no. Guests get in by QR code, NFC or link.
Can I start with one and add another later?
Yes. Starting with the most frequent tasks and opening up later is usually more orderly than launching everything at once.
What you can do this week
Fill in the table with the twenty requests your front desk gets most and mark which ones belong in conversation. If you want to test those answers with your information, you can create your agent and review the result with your team.
