01Step
Guest message arrives
From Booking.com, email, WhatsApp or the website chat — every channel lands in one inbox.
AI Guest Communication Platform
Hospitality
Recalibrated.
BaltIQ is the AI guest communication platform for hotels, hotel groups, serviced apartments and hospitality businesses. It answers guest messages across Booking.com, WhatsApp and email, reads your PMS before replying, and escalates to your team only when a human is truly needed.
- Replies in seconds
- 40+ languages
- Live PMS sync
- Human escalation
Unified InboxBooking.com · WhatsApp · Email
M. Lindqvistnow
Could we arrive after midnight?
Booking.comThe Haugens2m
Table for two, quiet corner?
WhatsAppA. Kalda11m
Is the sauna available at 07:00?
EmailBaltIQ drafting reply · English
ApproveRefineSend · 1.2s
Built by hoteliers
Designed from real hotel operations — focused on solving everyday guest communication.
- Written with front-desk teams, not for a demo stage
- Tested against real inboxes, PMS records and night shifts
- No public customer logos yet — we prefer honest proof
Chapter I
Four problems in
guest communication.
Hotels, hotel groups and serviced apartments all lose the same hours in the same four places. BaltIQ addresses them at the source.
I
Guest preferences are scattered
A quiet room, a dog, lactose-free breakfast — noted in an email, a PMS field, a sticky note. Nobody sees the whole guest.
II
The same questions, every single day
Check-in times, parking, breakfast hours, late arrivals. Staff retype the same answers hundreds of times a week.
III
Conversations live in four places
Booking.com messages, email, WhatsApp and the website chat never meet. Context is lost between channels and shifts.
IV
Revenue opportunities quietly disappear
Late check-out, transfers, spa slots, room upgrades — offered only when someone happens to have a free minute.
The principle
BaltIQ interrupts your team only when it matters.
Handled automatically
- Check-in and check-out times
- Parking, breakfast, Wi-Fi, directions
- Late arrival notes
- Availability and simple booking questions
- Upsell offers the hotel has approved
Escalated to staff
- Cancellations
- Refunds and payment disputes
- Complaints and service recovery
- Unusual or sensitive requests
- Anything the AI is not confident about
Chapter II · The WorkflowFrom message
From message
to answer.
Six steps, usually completed in under two seconds. Nothing is invented — every answer comes from your reservation data.
02Step
BaltIQ identifies the reservation
The message is matched to a booking, a room, a stay date and a guest profile.
03Step
It reads the PMS and the history
Reservation details, previous conversations and approved guest preferences are pulled together.
04Step
The request is analysed
Intent, urgency, language and sensitivity are assessed before a single word is written.
05Step
A reply is composed
Written in the guest's language and in your house tone, with the correct facts from the PMS.
06Step
Routine answered · important escalated
Everyday questions are sent automatically. Cancellations, refunds, complaints and unusual requests are handed to staff with full context.
Chapter III · The FlowOne line through
One line through
your whole operation.
Every guest message follows the same path — and every answer ends in Guest Memory, so the next conversation starts better informed than the last.
- 01
Guest
- 02
Booking.com · Email · WhatsApp
- 03
BaltIQ
- 04
PMS
- 05
AI analysis
- 06
Response
- 07
Guest Memory
- 08
Hotel staff — only when necessary
See it decide for itself.
Select a message. Watch BaltIQ answer it — or hand it to your team.
BaltIQ reply
Composing
“”
Confidence98%
SourcesPMS · Thread · Memory
Composed in1.4s
The core featureGuest Memory,
Guest Memory,
approved by you.
A returning guest books again. BaltIQ already knows she prefers a quiet room, travels with a dog, asks for extra pillows, needs lactose-free breakfast and writes in German.
But nothing is stored automatically. BaltIQ suggests a preference, your staff approve it, and only then does it become part of the permanent guest profile. Accurate, auditable, and always under hotel control.
Suggested
Detected from a real conversation, never invented
Approved
A member of staff confirms before it is saved
Applied
Used in every future reply, across every channel
Guest Memory · SuggestionsAwaiting staff approval
BaltIQ noticed three possible preferences for Ms. Lindqvist. Nothing is saved until a member of staff approves it.
- Approve · Dismiss
Prefers a quiet room
Detected in Email · 12 Mar
- Approve · Dismiss
Travels with a dog
Detected in WhatsApp · 12 Mar
- Approve · Dismiss
Lactose-free breakfast
Detected in Booking.com · 13 Mar
Approved preferences are used in every future reply. Dismissed ones are never stored.
Returning GuestsFollow-up cadence
Ms. Lindqvist
Last stay · 14 Mar- 3 months
A short note after a spring stay
Draft ready · awaiting approval
- 6 months
Sauna suite offered again, in German
Scheduled · staff approved
- 12 months
Anniversary of her first stay
Paused by hotel
Nothing is sent until the hotel enables the campaign. Every message is written from that guest's own history.
RetentionReturning guests,
Returning guests,
remembered well.
The guests you already know are the easiest ones to welcome back. BaltIQ keeps their history and, when the hotel chooses, writes a warm, personal follow-up three, six or twelve months after a stay.
Each message is written from that guest's own stay — the room they liked, the season they came in, the language they wrote in. This is relationship building, not a campaign blast.
Remembered
Every returning guest is recognised across channels
Configured
You choose the intervals: 3, 6 or 12 months
Personal
Written from previous stays and approved preferences
Controlled
Nothing is ever sent without the hotel enabling it
The ProductFour surfaces,
Four surfaces,
one guest record.
One inbox for every channel, an approved memory of every guest, a live line to your PMS, and plain numbers on what was automated and what was escalated.
Guest MemoryApproved profile
Ms. Lindqvist
Returning- Stays
- 4th visit · since 2021
- Room
- Quiet side · high floor
- Travels
- With a small dog
- Notes
- Lactose-free breakfast · extra pillows
- Language
- Prefers German
Every entry was suggested by BaltIQ and approved by hotel staff.
PMS SynchronisationBi-directional
Cloudbeds / Mews / Opera
Reservations · Profiles
Idle
Booking.com
Messages · Reservation data
Idle
Email · WhatsApp
Conversation history
Idle
Records reconcile every 40 seconds. No double bookings, no manual re-entry.
Automation FlowBooking confirmation
Message received · Booking.com
Reservation matched in the PMS
Approved late check-out offer composed
Confirmation sent · PMS updated
Four minutes of staff work, resolved in eleven seconds.
Communication AnalyticsRolling 12 months
76%
Messages automated
38s
Median first reply
24%
Escalated to staff
JanDec
Security
Careful with guest data, by default.
Guest data always belongs to the hotel. We store it in the European Union, encrypt it in transit and at rest, and never use it to train shared models.
Encrypted
In transit and at rest
Guest data
Stored in the EU, owned by the hotel
Permissions
Role-based access per property
Approval
Human sign-off before sensitive actions
- Cancellations, refunds and complaints always reach a human
- Every automated reply is logged and reviewable
- Preferences are only saved after staff approval
- Access can be limited per property, role and channel
Chapter IV · Connections
All connections →Guest ↔ BaltIQ ↔ PMS. Always in sync.
Cloudbeds
Mews
Opera
Apaleo
Booking.com
Email
New property management systems can be added through their APIs.
Chapter V · First house
Fonnental Design Hotel, Tallinn.
A design hotel in central Tallinn — our first pilot property. Projected figures below, to be replaced with measured results.
1,800
Messages answered / mo
62h
Staff hours saved / mo
€2,400
Additional services sold
Chapter VI · The Calculator
How much work could BaltIQ quietly handle?
Adjust a few simple values to estimate how many guest conversations BaltIQ can handle for your hotel, how many staff hours it can save, and what that work is worth each month.
BaltIQ performs work equivalent to a significant part of a front desk employee's workload, allowing hotels to grow without immediately hiring additional staff.
Guest messages / day60
Staff hourly cost€10.0
Handling time / message3 min
1,350
Messages automated / month
68h
Hours handled by BaltIQ / month
€675
Monthly labour value
≈ 0.4 FTE
Equivalent front desk employee
€1,458
Estimated upsell revenue / month
Estimates assume 75% of guest messages are routine and handled automatically, 160 working hours per full-time month, and conservative internal upsell assumptions. We refine these with you during a demo.
Chapter VII · How we build
We would rather show you the product than borrowed praise.
Built by hoteliers, not by a team guessing what a front desk does at 23:40.
Origin
Written inside real hotel operations
Nothing is invented. Every answer is grounded in the reservation, the thread and approved preferences.
Accuracy
PMS-grounded replies
The hotel keeps control. Sensitive situations, saved preferences and tone all remain your decision.
Control
Human approval by design
Chapter VIII · Investment
Starting from €350 per month.
Provisional pricing while we work with our first hotels, apartments and groups. It scales with the size of the property, not the number of messages.
Chapter IX · An InvitationA conversation,
A conversation,
to begin.
We work closely with a small number of properties while we build. Tell us about yours, and we will reply personally within one working day.
Talk to us
hello@baltiqnovum.com