Why travel personalization is broken

For the last two decades, travel technology companies have obsessed over search experiences. Filters, checkboxes, sliders, sort orders, star ratings, price bands, loyalty toggles, policy warnings — the list is endless.

The industry built remarkable infrastructure to retrieve travel options, but invested almost nothing in understanding travelers. The result: systems that know everything about inventory and almost nothing about the people booking a trip.

The myth of “more questions”

When travel companies talk about personalization, they usually mean taking an action based on a profile attribute. A window seat is automatically assigned based on a stated preference, or a seat upgrade is free because a traveler has loyalty status.

This approach quickly runs into limitations. Travelers are only willing to answer a limited number of questions, and their needs can vary for different trips.

If a traveler stayed at a business hotel last month, does that mean they want one today? If they redeemed loyalty points once, does that mean they always should? If they booked economy last year, does that reflect a preference or a constraint? 

Travel intent is situational, emotional, and contextual. 

A better approach: learning, not asking

At Spotnana, we’re looking at the problem differently. What if travel systems didn’t interrogate travelers, but learned from them one signal at a time?

Real preferences reveal themselves through the decisions travelers make throughout the booking process:

• When choosing between a boutique hotel and a global luxury brand.
• When deciding between the shortest flight and the best cabin.
• When picking a historic neighborhood vs. a beach resort.
• When choosing street food vs. a chef’s table.

These moments are where intent is real. Instead of interrupting the journey with questions, we need to let the journey teach the system.

From retrieval engines to learning systems

Traditional travel platforms are retrieval engines: “Here’s everything that matches your filters.” Spotnana is building a learning system: “Here’s what’s most likely right for you, and here’s why.”

Each choice a traveler makes will subtly reshape how our travel platform presents hotel rankings, flight prioritization, cabin selection, neighborhood clustering, and experience recommendations. Our system will adapt to the traveler, not the other way around.

A travel platform that learns a traveler’s preferences over dozens of trips becomes a fundamentally different tool than one that treats every booking as their first or puts travelers into generic buckets that may not match the intent of their next trip.

Privacy is paramount in our thinking and will be built-in by design. Our system will also allow customers to choose whether or not personalized offers are active.

Personalization within policy

In corporate travel, personalization has sometimes been seen as a source of conflict with travel policies. This framing is a mistake.

When travelers are forced to sift through results that don’t match their preferences, frustration grows, and the likelihood of leakage increases. 

By surfacing options that align with individual preferences within policy bounds, a learning system improves compliance and satisfaction at the same time. Personalization, when properly implemented, actually enables corporate travel programs to scale and operate more effectively than ever before.

Building systems that “know me”

The next generation of travel platforms will be able to observe, infer, and refine until they can anticipate what travelers need. This is a meaningful shift in how travel technology respects traveler time.

The architectural shift involved is significant. Instead of static profiles, backward-looking rules, and generic search results, travel platforms will need to implement living preference graphs, real-time learning, and intent-aware ranking. Once a travel platform does this well, the downstream effects can extend across offers, servicing, loyalty, upgrades, and disruption management.

Personalization needs to be driven by more than static preferences captured through a form. The travel industry needs to build systems that understand people, not just travel inventory.

This work is underway at Spotnana.