AI in hospitality, part 2 of 15
The last article looked at the five systems that run side by side in most hotel operations without talking to each other. The logical next question: do you need to replace them all with a single super-system?
No. And that is both the good news and the bad news.
What does "connecting systems" actually mean in a hotel operation?
Connecting systems does not mean replacing PMS, POS and CRM with a single piece of software. It means these systems automatically exchange information about the same guest — through defined interfaces (APIs) or a central data layer that serves as a shared reference point. The result is one reliable profile per guest instead of five contradictory partial views.
Why "replace everything" is the wrong approach
A PMS that has run reliably for years does not necessarily need to go just because it lacks AI connectivity out of the box. A complete replacement is expensive, risky, and ties up staff for months who are then missing from day-to-day operations. The real question is not "Which system do we replace?" but "Which data needs to come together so it forms a complete picture in one place?"
This is also the usual technical path: providers of customer data platforms for the hotel industry put exactly this function at the centre — bringing together guest data from PMS, POS, spa and booking engine into a single profile, without touching the source systems themselves (Ireckonu: Hotel Customer Data Platform).
Three principles instead of tool names
A list of software products is not very helpful here. More interesting are three principles that apply regardless of the provider chosen:
One guest profile, many sources. Each system can continue to do its own job — the PMS manages rooms, the POS records revenue. What matters is that all systems reference the same guest ID, so the information can be brought together in one place.
Unified master data. Name, contact details, preferences and special requests should be maintained in one place only and distributed from there to all systems — not entered by hand five times over. This is exactly where most errors happen in practice: duplicate entries, outdated phone numbers, contradictory notes.
Consider data protection from the start. Anyone merging data also needs to clarify who may access what, and how long information is retained. This belongs in the planning from the beginning, not as an afterthought for compliance — precisely because a merged guest profile reveals more about a person than any single system on its own.
An example from everyday practice
A four-star city hotel with a restaurant and a small spa area long had the same picture described in the last article: the PMS knew the room preference, the POS knew the restaurant preferences, the spa software had separate bookings. At check-in, reception often did not know that a guest had booked a massage the evening before.
The solution was not a new PMS. Instead, the three systems were connected through a central data layer that brings all information together under a single guest ID. Since then, reception sees a complete picture at check-in: room preference, upcoming spa appointments, known intolerances. The systems themselves remained unchanged — only the connection between them was new.
This connection is the first automated process in this series: a rule that, without manual intervention, ensures information reaches where it is needed. Later articles will cover more complex processes on the same basis — for example in staff planning or revenue management.
Linking back to the data foundation
The benefit is not limited to the front desk. Once a unified guest profile exists, an AI assistant in marketing can draw on the same data basis as the team at reception. Newsletters, website personalisation and guest communication access the same, current information — instead of a data export that was already outdated by the time it was created. That is the difference between a marketing tool working in isolation and one built on a solid foundation.
An exercise for your own operation
Ask yourself: if a guest calls today asking about their next reservation, how many systems does a staff member need to open to gather all the relevant information? If the answer is more than one, there is a seam here worth a closer look.
What comes next in the series
The next article covers the AI status assessment: a structured process for finding out where your own operation actually stands in terms of systems, data, team and processes — and which measures pay off over which time horizon.
Related articles
- The data graveyard most hotels are sitting on (part 1 of the series)
- Requirements for a booking engine on the hotel website
- Hyper-personalisation 2.0: when the hotel website responds individually to each guest
Sources
- Ireckonu (2026): Hotel Customer Data Platform (CDP)
- Revinate (2026): 2026 Hospitality Benchmark Report / Hotel Database Health
Frequently asked questions
Does a hotel need to replace its PMS to connect systems?
No. In most cases an interface or a central data layer that connects existing systems through a shared guest ID is enough. A complete replacement is rarely necessary and usually not the sensible option either.
What is a unified guest profile?
One data record per guest that brings together information from several systems (PMS, POS, CRM, spa), so every department has the same, current view of preferences and booking history.
Where should a hotel start with connecting systems?
Best to start where the disconnect is felt most strongly in daily operations — often at the interface between reception and ancillary operations such as restaurant or spa. A structured stocktaking shows where the biggest gaps are and what should be tackled first.
Tourismusconsult supports hotel operations with the AI status assessment: a structured analysis of systems, data, team and processes that leads to a prioritised roadmap — with quick wins, medium-term and strategic measures.
