How hospitality leaders in Asia Pacific are turning data into decisions, culture, and competitive edge

There’s a particular kind of silence that falls in a boardroom when a chart contradicts every assumption in the room. It’s the silence of a story being rewritten in real time. That moment — when a single number shifts perspective — was the starting point for a recent roundtable on data storytelling in Asia Pacific hospitality, bringing together commercial leaders grappling with one of the industry’s most persistent challenges: not just collecting data, but making it mean something.

What followed was a candid, practical, and sometimes uncomfortable conversation about the gap between having information and actually using it.

First, Assume the Data Is Wrong

The most counterintuitive piece of advice from the discussion came early: before you tell a story with data, question whether the data is telling you the truth.

One participant described a striking example from a luxury hotel where analytics flagged Afghanistan as the top source market for international guests — a result that would have sent a marketing team chasing an audience that simply wasn’t there. The real culprit? Front office staff who hadn’t been properly trained to complete nationality fields, resulting in a cascade of dirty data that distorted the entire picture.

“Always question data first,” was the consensus. “Assume it’s wrong until it’s verified.”

The sources of distortion are more varied than most teams realize. VPN usage causes guests to appear as if they’re booking from entirely different geographies. Declining cookies creates invisible gaps in tracking. And in markets where digital privacy awareness is growing, those blind spots are getting bigger.

The practical response isn’t paralysis — it’s double-sourcing. The roundtable pointed to a simple but underused tactic: include “How did you hear about us?” fields in booking forms and guest surveys, then cross-reference those answers with operations teams who interact with guests directly. Human intelligence, it turns out, is still one of the best validators of algorithmic output.

AI as a Second Pair of Eyes — Not a Replacement for Judgment

AI-powered anomaly detection was embraced by the group, but with deliberate constraints. The advice was pointed: upload anonymous screenshots, not files. Uploading raw data files risks metadata leakage and creates unnecessary exposure. Anonymous screenshots preserve the insight while protecting the underlying data and its confidentiality.

The more interesting application of AI in this context isn’t generating headlines — it’s surfacing the quiet signals that human analysts miss. A one or two percent variation in a metric rarely triggers alarm, but compounded over time or across segments, it can signal something significant. Asking an AI tool to specifically hunt for those small, sustained anomalies adds a layer of pattern recognition that complements rather than replaces human analysis.

The broader question of AI’s role in the storytelling layer generated real debate. As AI-generated insights become standard features in commercial strategy platforms, who is accountable for the narrative those insights create? The concern isn’t that AI will get it wrong — it’s that teams might stop asking why, accepting the algorithm’s framing without interrogating its assumptions.

The group’s view: human judgment must stay in the storytelling layer. AI can identify what’s happening. The “why it matters” still belongs to people.

Revenue Management Is More Nuanced Than One Model Fits All

On the commercial strategy side, the discussion dismantled the idea that there’s a universal formula for ADR versus occupancy trade-offs.

For city hotels, the math often tilts toward pushing average daily rate. When guests are in town for a conference or a business meeting, they’re unlikely to spend heavily on ancillary services — spa, dining, leisure. Protecting rate makes sense.

Resort properties tell a different story. An unoccupied room doesn’t just mean lost accommodation revenue; it means a guest who isn’t spending at the restaurant, the pool bar, the wellness centre. The calculation has to account for that missing downstream spend.

All-inclusive resorts bring the equation back around again — with food, beverage, and activities bundled into the rate, the focus returns to rate optimization itself.

More broadly, the group pointed to a shift in how the most sophisticated commercial teams are thinking about value. Segment-based pricing is giving way to individual guest scoring — calculating the full lifetime value of a guest as (stays × spend × retention years) minus acquisition cost. It’s a more demanding model, but it reflects a more honest picture of who is actually valuable to a property over time.

Average length of stay data is another leading indicator worth watching closely. Recent trends show guests preferring shorter stays than in previous years. That shift in behaviour has real implications for revenue strategy, staffing, and programming — but only if teams go beyond noting the trend to investigating the cause.

The Art of Presenting Uncomfortable Findings

Perhaps the most practically useful part of the conversation concerned what to do when the data tells leadership something they don’t want to hear.

The key, participants agreed, is to disassociate yourself from the finding. “I was surprised to see…” does real work as an opener. It frames the discovery as a shared revelation rather than a personal challenge to anyone’s decisions. It also positions the analyst as someone who came to the data without an agenda — which, over time, builds the kind of credibility that makes difficult conversations easier.

Equally important: never present a problem without at least two paths forward. Offering Plan A and Plan B gives leadership agency. They’re not being told what went wrong; they’re being invited to choose how to respond. Data becomes a tool for decision-making rather than a verdict.

The challenge is compounded by the fact that different departments don’t actually measure success the same way. Revenue teams track bookings and conversion rates. Marketing lives and dies by ROAS — return on ad spend. Leadership tends to care most about overall profitability. Presenting the same data to all three audiences with identical framing is a recipe for each of them to hear something different.

The investment in brand building versus lead generation is one area where this misalignment shows up acutely. Brand spend is long-term by nature — no one expects immediate ROI from awareness campaigns. Lead generation, by contrast, should show measurable returns within three to six months. Treating both with the same measurement framework, or failing to set expectations upfront, sets up conflict that data alone won’t resolve.

Building a Culture Where Data Actually Gets Used

Getting teams with different skills, priorities, and levels of comfort to trust and act on the same data is, ultimately, not a technology problem. It’s a culture problem.

Data literacy in hospitality means different things at different levels of an organisation. A front office manager needs to understand why it matters that they fill in every field correctly. A revenue manager needs to be able to explain a pricing decision to a general manager who may not share their analytical framework. A marketing lead needs to hold two things in tension: the long-term value of brand investment and the short-term pressure to show returns.

The roundtable offered no single answer to how you build that culture — because there isn’t one. But the consistent thread was that data credibility is earned through the stories told around the data, not the dashboards that display it.

A well-designed dashboard that no one believes is just noise. One that’s been verified, contextualised, and explained in terms of what it means for the business — that’s what actually changes decisions.

_______________________________________________________

This article is drawn from a roundtable discussion at HSMAI APAC Commercial Strategy Conference 2026 on 15-May at Marina Bay Sands, Singapore.