Closing the personalization gap: Why hotels still struggle to connect online promises with offline reality

A guest books a hotel room on their phone. The loyalty app remembers their preferences based on loyalty data, prior bookings, and browsing behavior. It makes suggestions that feel intuitive and personalized: a high floor, a spa treatment, a late check-in. It even suggests suitable dining options and local activities. From time to time, it surprises them with an upgrade or a suggestion for a local experience they did not expect, but that makes sense, all based on transaction data, look-alikes and past browsing behavior. The experience feels seamless, intuitive, almost like the hotel knows them better than they know themselves.

Hotels have spent years perfecting this digital personalization, and GenAI amplified it. We have built sophisticated booking engines, smart pricing models, dynamic email campaigns, and loyalty apps that know our guests better than they know themselves. But somehow, when that same guest walks into the lobby, all that intelligence vanishes.

The online journey feels tailored and intuitive while the offline experience can often feel scripted and underwhelming. It is the great personalization paradox of our industry, and it persists not because we lack data or intent, but because we have failed to connect the two worlds. And this disconnect is not anecdotal. According to Deloitte, nearly 60 percent of travelers expect personalized experiences across the full journey, yet fewer than 25 percent feel hotels deliver consistently once they are on property.

Why the disconnect exists

The promise of personalization has always been to make every guest feel known, understood, and valued, yet there are only a few hotels that do this right, and they often rely on manual processes like looking at photos of VIPs during the morning briefing.

The reasons for this lack of automated personalization are structural as much as technological. Many hotel groups operate under franchise or management contracts: meaning, the brand controls the data but the property delivers the experience. Integrating data from CRS, Loyalty, CRM, PMS and personalization engines is complex, and front-line staff often lack the tools, training, or incentive to act on guest insights.

Even if a brand manages to integrate all its data seamlessly, it is not often turned into actionable information for front line staff. In addition, there are not many operational tools or apps available that can present this information in a way that is actionable for busy hotel operations.

Online personalization thrives because it is rule-based and structured: an algorithm matches preferences with offers.

In contrast, offline personalization depends on humans, and people rely on access, tools, training, and timing.

The front desk might know the guest’s loyalty status but not their last three feedback scores. The restaurant might be aware of dietary preferences but not arrival times. Housekeeping may see “VIP” in the notes, but not why. Often, each department has its own slice of the truth, but no one sees the whole picture.

The human factor

Technology alone cannot fix this gap. People deliver hospitality, not code. The best CRM in the world means nothing if staff don not understand what to do with it or it is too complex to use in a demanding hotel environment.

In many hotels, personalization has been delegated to the marketing team. It is measured in open rates and conversion scores rather than emotional impact. But true personalization is a cultural discipline, not a marketing feature. It requires that everyone, from revenue managers to room attendants, share the same mindset and use data to care better, not just to sell better.

Training, empowerment, and recognition are essential. When employees are encouraged to act on insights and trusted to make small, spontaneous gestures, personalization becomes part of the brand’s DNA.

A great stay rarely comes from perfect software. It comes from the moment someone says, Welcome back, we kept you on the quiet side of the building like last time.

I once worked with a brilliant marketeer, who was obsessed with creating structured special moments for our guests. Together, we designed an algorithm to rank customers and assign them special moments once they stayed in one of our hotels. The on-property staff would get triggers via the CRM system showing them simple steps on what to do, what to say and when to do it to create those personalized moments. The prototypes worked very well but, due to technical limitations, it never scaled, and we eventually had to abandon the project.

Why new technology might finally bridge the gap

So why might it be different this time? How can emerging technology solve a problem we have been debating for decades?

It is because the next generation of systems are not just about collecting data: they are very good at connecting them, even if they live in different environments and may not be perfectly structured. Below are the four drivers:

Unified data layers and the “living profile”

New open-architecture PMS and CRM platforms are starting to build what we have long talked about: a single, real-time guest profile that moves across the entire journey. Instead of static preferences, these profiles behave like living organisms, constantly learning from interactions, purchases, and sentiment signals like reviews and surveys.

AI-driven context and prediction

Artificial intelligence now enables systems to understand guest intent from subtle cues: not just what someone booked, but why they did. By analyzing tone in messages, stay frequency, or even similar guests, AI can predict what a guest might value next: a quiet table, a late checkout, or a personalized recommendation nearby.

Agentic and embedded intelligence

We are entering the era of agentic AI - autonomous systems that respond to input and act on behalf of both guests and staff. Imagine a digital concierge that can coordinate with the PMS, housekeeping, and the restaurant to anticipate a guest’s request before they make it, or a staff co-pilot that quietly reminds an associate that a repeat guest prefers sparkling water over still.

Seamless context transfer

The Model Context Protocol (MCP), a new open-standard initiative, will soon allow hospitality systems to share relevant guest context securely across platforms. In practice, this means a guest's preferences captured in a chatbot could guide how they are greeted on arrival, or what offers appear on their app.

From data to moments

But let us be clear, data does not create loyalty, as my colleague always said, special moments do. Technology may make personalization possible, but people will make it memorable. We don not need more data or another app or widget, the question should be: How can we use what we know to surprise, reassure, or delight this guest right now?

Hospitality has always been about connection, the warmth of being recognized, the small details that say, We were expecting you. For too long, technology has automated the transaction but ignored the emotion.

We are now at an inflection point. The convergence of unified data, AI, and open protocols gives us the tools to make personalization finally real; online and offline, digital and human.

If my colleague from way back is reading this; I hope you have not given up. You were ahead of your time. It is only now that we can make AI enabled personalization on property a reality at scale. Let us see who makes it happen first!