AI automation for hospitality the front desk goes back to the guest
Bookings land from your own site, from booking portals, from the phone and from email, and the guest is already asking about parking, check-in and breakfast before they arrive. We pull those channels into one place: repeat questions get an answer from your property knowledge base, and occupancy and revenue for today and tomorrow sit in a single view.
- 14 daysto the first working automation, counted from the end of the workshop
- EUR 3,500 netper week of workshop, a published price we do not negotiate
- 5 to 10 secondsthat is how long the AI agent takes to answer a guest, from your property knowledge base
The problem
Why does the front desk answer email instead of the guest at the counter?
A guest stands at the counter waiting for a key. In the inbox sit fifteen messages asking the same things: is there parking, when is check-in, how late is breakfast, are dogs allowed, how do I get in from the ring road. Every one of those answers already exists, just not in one place: once in the house rules, once in a listing description, once in the head of the front desk manager.
Bookings arrive by four routes at once: your own site, booking portals, the phone and email. Each route leaves the data in a different shape, and whoever is on shift stitches the day together out of them. So the question of how many rooms are taken tomorrow and what revenue that is means logging into several places and a quarter of an hour in a spreadsheet, and the answer holds until the next booking.
Guest data spread across an inbox, a spreadsheet and notes at the desk is a formal risk as well. The GDPR (Regulation 2016/679) gives you one month from receiving the request to answer a guest who asks what you hold on them or asks you to erase it (Article 12(3)). When that data sits in five places, the answer is a reconstruction from memory rather than a printout from a system.
- The receptionist types the parking answer out again instead of serving the guest at the counter
- A group enquiry sits in the inbox until the afternoon, because two details are missing from the quote
- The occupancy and revenue figure for tomorrow is assembled by hand, so it happens once a day or not at all
- A guest request raised on the morning shift disappears at the handover to the afternoon
Why now
What changed, so this pays for itself today?
Property management systems and channel managers now have interfaces that let you read bookings without exporting to a spreadsheet. What used to be a quarter-long project three years ago is now counted in working days for one person.
The second change matters more. A rule will copy a booking from a channel into your system, but it will not answer a guest who asks at 23:00, in their own words, whether they will still make breakfast after a late arrival. The AI agent reads your property knowledge base and replies in a sentence within 5 to 10 seconds, in the channel the guest wrote to.
The third thing is about the team. In a seasonal property part of the staff works a few months and never gets to learn the unwritten rules. A written property knowledge base and a task flow for housekeeping turn that knowledge into something you can hand over on day one, instead of explaining it in the corridor when there is a moment.
What changes
What does a property have after the first month?
Pre-stay questions have one address
Parking, check-in and check-out, breakfast, pets, directions. The guest gets an answer from your property knowledge base, in the channel they wrote to, and the front desk only sees what needs a decision.
An answer in 5 to 10 seconds, including outside front desk hours
Confirmations and pre-arrival messages send themselves
Booking confirmation, directions and parking instructions, a reminder of the check-in time. Sending is tied to the arrival date, not to a free moment at the desk.
Directions go out 2 days before arrival, with nobody clicking send
Occupancy and revenue in one view
Bookings from every channel come together into a single view for today and tomorrow, split by source. Nobody assembles that from several screens in the morning.
One view instead of a quarter of an hour in a spreadsheet, refreshed after every booking
Group enquiries arrive complete
The form and the AI agent collect the dates, the number of guests, the need for meeting space and catering before the enquiry reaches whoever writes the quote. The email thread chasing two missing details disappears.
Everything needed for a quote in the first message, not the fourth
Housekeeping and maintenance in one flow
A report from the front desk becomes a task with a room number, a priority and an owner. Status is visible without phoning the floor, and a shift hands over a list instead of notes.
A task with a status instead of a verbal handover between shifts
Post-stay survey and ratings in one place
The request for feedback goes out after check-out, and the replies and the ratings from outside channels land in one view. You see what repeats, not the single comment somebody happened to read.
Recurring remarks counted, not remembered
Proof
Do you use this yourselves?
We do not have a rollout in a hotel whose numbers we are allowed to publish, and we are not going to get around that with an anonymous example. So we show the only thing we can show in full: an AI agent running in production in our founders' own company, built with the same method. Together with what those numbers do NOT prove.
Built for ourselves. These are not client deployments but implementations inside the founders' own companies. We say so plainly, because proof you cannot verify is not proof.
JustBottheir agentJustIdea Agency, a performance marketing agency of around 30 people
- 23
- people from the company work with the agent
- 38
- systems wired in permanently
- 112
- processes run without a human
- 475
- thousand lines of the agent’s code
The agent is wired into the task system, the team calendars, the CRM and the data warehouse. First commit 1 March 2026, 23 people use it today. It answers organisational questions: who is available, what is happening with a case, what came in from a form.
8 of 38 systems it works in
Google Ads
Meta Ads- HubSpot
- ActiveCollab
- Ahrefs
GA4
Search Console
BigQuery
- Before
- Questions about who was free and where a case stood circulated between people. The answer depended on who happened to have the right system open and a spare moment.
- After
- The same questions go to the agent and come back from data, in seconds. A person steps in where something has to be decided, not where something has to be looked up.
8,631 operations on team availability and 1,301 on form enquiries in 30 days
And in your property
What if things work differently here?
Four sentences come up in accommodation businesses more often than any others. We answer them here the way we answer them on a call.
We have a property management system and a channel manager we are not going to replace.
There is no need to. We read the data out of them through the interface and leave it where it is. Replacing a system in a property that is open for business is a project of many months, and here the point is 14 days to the first result.
Nothing about our property is written down anywhere.
It almost never is, and that is the first thing the workshop produces. Check-in times, the pet policy, parking, directions, breakfast hours and the exceptions you make for regulars go into one base. Only once it is written down can it be served to a guest automatically, and only then does a new person at the desk stop guessing.
Our guests will not talk to a machine.
Some of them will not, and that is fine. The AI agent takes the repeat questions and passes everything else to a person, along with the whole conversation. A guest asking about parking at 23:00 gets the answer straight away instead of waiting until morning.
We have 24 rooms, that is too small for this to pay off.
In a small property the gain sits somewhere else than in a chain. Not in reporting, but in one person on shift no longer typing the same answers and writing confirmations by hand. In the workshop we say which of those effects is worth money in your case, and if none is, we say that too.
- Bookings and availability
- property management system, channel manager, booking portals, the form on your site
- Guest contact
- email, phone, SMS, messaging apps, chat on your site
- Teamwork
- Google Workspace, Microsoft 365, Slack, task lists
- Billing and documents
- your invoicing service, an on-premise accounting program reached through a connector inside your network, document flow
- Data and reports
- Google Sheets, BigQuery, SQL databases, revenue spreadsheets
- Automation layer
- Make, n8n, custom API integrations
What exactly we automate in an accommodation business
A typical split of front desk work before a rollout and after it. The order we take it in is agreed in the workshop.
| Area | Today | After the rollout |
|---|---|---|
| Pre-stay questions | the front desk answers by hand in each channel separately | an answer from the property knowledge base, in the guest's channel |
| Bookings from several channels | retyped between the system, email and a spreadsheet | one place to save them and one shape of data |
| Pre-arrival messages | sent when somebody has a moment | sent automatically, tied to the arrival date |
| Occupancy and revenue | assembled in the morning from several screens | one view for today and tomorrow, split by channel |
| Group and event enquiries | an email thread chasing missing details | everything needed for a quote in the first enquiry |
| Housekeeping and maintenance tasks | a note at the desk and a call to the floor | a task with a room, a priority and a status |
| Post-stay feedback | requests sent irregularly, ratings read separately | a survey after check-out, ratings in one view |
A first rollout usually covers three lines from this list, not all of them at once. We pick the ones that take the most work off the front desk.
How we work
Four steps from the front desk inbox to a flow that runs
Every step ends with something you can judge before you decide on the next one.
Workshop on site
One week. We walk the guest route from the first enquiry, through booking and the stay, to check-out and the invoice. We come out with a map and a list of the places where data is created twice or lost between shifts.
Picking three automations
From the map we take the three with the highest return and measure them before the start: how many pre-stay questions come back in a week, how long the occupancy report takes to build, how many group enquiries wait longer than a day.
First rollout in 14 days
In an accommodation business we usually start with pre-stay questions, because those take the most work off the front desk. The property knowledge base is written once and serves every contact channel.
Measurement and the next areas
After 30 and 90 days we check the indicators agreed in the workshop. Only then do we decide whether to go for the occupancy view, for group enquiries, or for the housekeeping task flow.
Before and after
How the day of a front desk manager changes
Before
- Mornings go on retyping bookings from portals and checking nothing was double sold
- Fifteen messages about parking, check-in and breakfast get answered by hand
- Tomorrow's occupancy is assembled from several screens for an email to the owner
- A guest request is passed to the afternoon shift verbally, along with the keys
After
- Bookings from every channel are in one place before the laptop is even open
- Repeat questions are answered without him, and he steps into the exceptions
- The view for today and tomorrow is waiting, split by channel
- A guest request is a task with a status, not something said out loud
The first effect shows on the day pre-stay answers go live. The occupancy view and the channel mix we judge after 30 and 90 days, on indicators set before the start.
For the owner and the front desk manager
What the person signing this off gets
The first thing is visible straight away: the front desk stops being an inbox and goes back to the guest standing at the counter. That is the work you are paying for, and the only part of a stay people can really shape.
The second is quieter, and it counts for more in a conversation with an owner or a bank. Occupancy and revenue for tomorrow stop being the knowledge of whoever is on shift and become a number you can show in the same shape every day. What the property knows stays in the business, even when the seasonal team turns over.
- A property knowledge base written once and used by every channel
- One occupancy and revenue view, in the same shape every day
- A new person at the desk starts with ready answers, not a notebook
- Indicators to measure after 30 and 90 days, agreed before the rollout
What it costs
The workshop price is published, the rollout price comes after it
The workshop costs EUR 3,500 net per week and we publish that before any call, because a property has a right to know what it is signing up for. For a single property one to two weeks is usually enough. We quote the rollout after the workshop, once we know how many channels and systems we are connecting and what state they are in.
You pay for the language model and the automation tools directly to the vendor, with no commission from us. A model query costs the provider roughly EUR 0.005 to 0.07. If the accounting program runs on a machine inside your building, we reach it through a connector on your side, so the data does not travel where it should not.
Process workshop
EUR 3,500 net
per week, in an accommodation business usually one to two
A self-contained piece of work. It ends with a process map, a written property knowledge base and a 30 day plan you can execute without us.
- A week of work on site and in your systems
- A map of the guest route from enquiry to check-out and invoice
- Three automations with the highest return, measured before the start
- The indicators to measure after 30 and 90 days
- Most popular
Workshop and first rollout
Quoted after the workshop
scope follows from the number of channels and the state of the systems
The variant most properties choose. Answers to pre-stay questions are live within 14 days of the workshop ending.
- Everything in the workshop
- The first automation live in 14 days
- Connections to the property management system and the guest contact channels
- Training for the person who runs this on your side
- Measurement after 30 and 90 days
AI agent for a property
Quoted after the workshop
for properties where the same questions come back every day
Everything in the rollout plus BOND, the company AI agent that answers questions from guests and staff out of your own data.
- Everything in the first rollout
- An AI agent answering from the property knowledge base in 5 to 10 seconds
- Staff questions about occupancy, revenue and room status in plain language
- Access for the front desk, housekeeping and the owner
If after the workshop we conclude there is nothing here worth automating, we say so plainly and do not propose a rollout. The workshop still ends with a document that holds without us: a process map, a written property knowledge base and a 30 day plan you can carry out with your own hands.
Frequently asked questions
Will an AI agent replace the front desk in a hotel?
No, and that is not why we build it. Automation will not replace a front desk or a host. It takes over repeat questions and reporting so that people can look after the guest standing at the counter. Anything outside the written property knowledge base is passed to a person, together with the whole conversation.
Where does the AI agent get its answers for guests?
Only from the property knowledge base we write down in the workshop: check-in and check-out times, parking, breakfast, the pet policy, directions, facilities, cancellation rules. Nothing is invented. A question outside the base gets an answer saying it is being passed to the front desk, and it reaches a person with the full context.
Do we have to replace our property management system or channel manager?
No. We read the data out of them through the interface and leave it where it is. Replacing a system in a property that is open for business is a project of many months and it lands in the middle of the season, while the first working automation is meant to be running 14 days after the workshop ends.
How long does a rollout take in an accommodation business?
One week of workshop and 14 days to the first working automation, counted from the end of it. A fuller scope, with the occupancy view and the housekeeping task flow, usually fits in 2 to 4 weeks. We measure the effect after 30 and 90 days, on indicators agreed before the start.
Will this work in a seasonal property with 20 rooms?
It will, only the gain sits somewhere else than in a large hotel. In a small property the saving on reporting is slight, and the real effect comes from answering pre-stay questions and sending confirmations without a person. In a seasonal property there is a second benefit: a new person on shift has the rules written down from day one.
Will automation raise our occupancy and our guest ratings?
We do not promise that and we cannot guarantee it. We promise two things you can check: the guest gets an answer faster, in 5 to 10 seconds instead of hours later, and occupancy and revenue from every channel are visible in one place. What you do with a faster answer and with those numbers is up to you.
What about guest data and the GDPR?
The data scope, access per role and where things are stored are agreed in the workshop and written down before launch. Answering pre-stay questions usually needs no personal data beyond a first name and a booking number. The data processing agreement is signed before the start, not after it.
How do group and event enquiries work?
The enquiry collects everything a quote needs: the dates, the number of guests, the number of rooms, the need for meeting space and catering, the budget. It lands in one place with a response deadline and a reminder, instead of sitting in an email thread. Whoever writes the quote gets a complete set rather than three things to chase.
Let us start with the questions that come back every day
The free consultation takes 30 minutes. You tell us how many rooms the property has, which channels bookings arrive through and what guests ask most often before arrival. We tell you what can come off the front desk in 14 days and what needs the property rules written down first. You do not have to give us access to any system at this stage.
We run at most two rollouts at a time, so that someone is actually sitting on each of them.
Book a free consultationPS. If somebody on your team answered the parking question again this week, that is exactly the problem. The answer exists, it is just not written down in one place, so a person types it out instead of serving the guest at the counter. Writing it down is where the workshop starts.