Personalisation That Pays: The Version Hotels Actually Need
Everyone's selling hotels personalised guest experiences. Almost none of them ask whether your team's own systems understand your hotel first, and that gap is where the real revenue is leaking.
Every hotel conference has the same slide by now. A guest walks in, the system recognises them, a bottle of their favourite wine is already chilling, the app knows they like a late checkout. Personalisation, the industry says, is the next frontier.
Ask yourself honestly: has any of that moved your RevPAR this year?
For most hotels, the answer is no. Not because the guest-facing vision is wrong; it's a nice ambition. It's because it solves the wrong problem first. Before a hotel can personalise anything for a guest, someone on the team has to notice the opportunity, understand why it matters, and act on it in time. That's the part nobody's automating. And it's the part that's actually costing hotels money.
What Hotel Personalisation Actually Means
Ask most vendors what "personalisation" means and you'll get the same answer: guest recognition, tailored offers, upsells triggered by past stays. That's personalisation aimed outward, at the guest.
There's a version aimed inward that almost nobody talks about: a system personalised to how your specific hotel runs: your segments, your comp set, your rate floors, your team's rhythm. That's the version that actually protects revenue, because it's the version that decides whether anyone notices the opportunity in the first place.
Why Guest-Facing Personalisation Isn't Moving RevPAR
Here's what personalisation looks like in most properties today: a revenue manager opens Opera. Then IDeaS. Then a spreadsheet for comp set rates. Then last year's numbers, pulled separately, because none of these systems talk to each other.
"Right now I have 11 tabs open. I need to log into Opera, run the report... it takes me 45 minutes just to see where we are."
That's not a hypothetical. That's a revenue manager describing her actual Tuesday morning.
None of those 11 tabs are personalised to her. They're generic; same report, same layout, same blind spots, whether she's running one property or three. Her RMS doesn't know her portfolio is three sites, not one; it treats every property like it's on its own island. Her PMS won't tell her a block just washed down unless she goes looking for it herself.
A guest-facing chatbot doesn't touch any of that. It sits on top of a system that's already behind.
The Personalisation Gap Hotels Are Actually Missing
The real gap isn't "does the guest get a birthday email." It's: does the system understand your hotel well enough to tell you what matters before you have to go find it?
That gap shows up every morning, in the same shape, at property after property; reports that require manual stitching, alerts that arrive too late to act on, and RMS recommendations that sit unused because nobody had the hour to act on them.
What Personalisation That Pays Actually Looks Like
There's a story worth sitting with. The Alex, a 103-room hotel in Dublin's O'Callaghan Collection, had a revenue team with real skill and a strong RMS in IDeaS, but IDeaS sets rates, it doesn't tell you where to override it. That judgement call was taking an hour a day minimum, and there were never enough hours to cover every date. So the team handed the daily analysis to Otel: every morning, before anyone walked in, it reviewed the full forward calendar and flagged where the hotel was underpriced against the comp set. The revenue manager still made every call, they just spent the hour acting on it instead of gathering it.
Over three months, the team shipped about 120 rate actions a month. RevPAR grew 8.6% year-over-year — 10% above the comp set average — with ADR up 4.6% blended across the period. Full breakdown →
The RMS was always right. The team just never had the hour back in the day to act on it. That's what personalisation that pays looks like: not a system personalised to the guest walking through the door, but one personalised to how your revenue manager, your GM, and your owner actually need to see the day:
- Reports built around your segments, not a template.
- Alerts timed to when you can act on them, not a static 8am email nobody reads until 11.
- Reasoning shown in plain language, so a €10 rate drop isn't "the algorithm said so", it's tied to a 3-day pickup pattern and what worked last time the comp set moved the same way.
Real Results: How Operational Personalisation Protects Revenue
A few more examples of what that has uncovered, quietly, in the background:
- €3,500 in room revenue that had gone missing overnight: a guest checked out early, nobody flagged it, several people looked and couldn't find it. Fifteen minutes, once someone knew where to look.
- 83 rate code violations in a single month: bookings that slipped in below the floor rate because nothing was watching for it.
- A menu item costing more to produce than it was sold for: caught in the first week of looking at F&B margins properly.
- €900 a month in kitchen wastage: hiding in a teaspoon-level cost variance nobody had ever pulled apart.
None of that is guest personalisation in the conference-slide sense. It's personalisation in the sense that actually protects revenue: the system knows your hotel's specific patterns well enough to flag what's off before it compounds.
Why Guest Personalisation Should Come Second
This isn't an argument against guest personalisation. Understanding a guest well enough to offer the right upgrade at the right moment matters too. But sequencing matters. You can't personalise the guest experience with a team that's still spending its mornings stitching together reports from five disconnected systems.
The operational foundation has to come first: one place where your hotel's data lives together, understood in your hotel's terms, surfaced to the right person before the moment passes. Get that right, and the guest-facing layer becomes a lot more achievable, because the team finally has the hour back to think about the guest at all, instead of the report.
Questions to Ask Before Buying "Personalised" Hotel AI
Does it personalise your team's view of the hotel, or just the guest's? Most platforms personalise outward. Ask what it does with your internal data: your segments, your reporting rhythm, your team's decisions.
Does it know the difference between a compression night and a Tuesday? If the alerts and reports look the same regardless of your calendar, it isn't personalised to your property.
Does it know your comp set and rate floors specifically, or is every property on the call getting the same dashboard? Generic dashboards feel personalised because they're customisable. That's not the same as a system that already understands your hotel's patterns.
The Bottom Line
The hotels getting the most out of AI right now aren't chasing the flashiest guest-facing feature. They're the ones who gave their revenue managers and GMs a system that understands their property specifically, and got an hour of their morning back to act on what it finds.
That's personalisation that pays. Everything else is a nice demo.
Continue reading
Related articles
Vibe Coding and Hotel Operations: What It Means for Your Team
Vibe coding isn't just a developer trend. It's for hotel ops, letting revenue managers and GMs describe what they need in plain English and get it done.
Paul Ryan·6 Jul 2026
Hotel AI Skills Gap 2026: Why Technology Isn't the Problem
Most hotels have the tools. What they're missing is the capacity to act on their data. Here's what the AI research from BCG says, and what to do about it.
Paul Ryan·24 Jun 2026
Your Hotel Systems Are Already Collecting The Data. Otel AI Connects Them.
Otel AI connects Opera, IDeaS, Lighthouse, and your other hotel systems into one platform. Set up Flows in plain English. Intelligent analysis, reports and alerts arrive in your inbox automatically.
Paul Ryan·11 Jun 2026