How Singapore Earned S$23.9 Billion by Ignoring the Number Every Tourism Board Obsesses Over
Created on 2026-03-14 12:21
Published on 2026-03-15 12:45
Singapore missed its target.
In the first nine months of 2025, the city-state welcomed 16.9 million international visitors. The floor target was 17 million. By every traditional measure of tourism success, the Singapore Tourism Board fell short.
Then the revenue numbers came in. S$23.9 billion. A national record. Up 6.5% year-on-year.
Fewer visitors. More money. By a lot.
The average per-visitor spend hit S$1,804. Per-day yield climbed to S$372. Food & beverage spending surged 15%. Sightseeing and entertainment, also up 15%. Singapore didn’t just tolerate the shortfall in arrivals. It had engineered for it.
This is the story most people miss when they talk about Singapore’s tourism success. They credit the Marina Bay Sands expansion. The Formula 1 night race. The new Mandai Wildlife Reserve. These matter. But they’re outputs.
The input? A data platform called Stan that most people outside the tourism industry have never heard of. And an organizational approach to AI readiness that every city in Asia-Pacific should be studying.
The Platform Nobody Talks About
The Singapore Tourism Analytics Network, known as Stan, is not a dashboard.
I need to be specific about this, because “data platform” has become one of those phrases that means everything and nothing. Every tourism board claims to have one. Most of them are spreadsheets with a login page.
Stan is a multi-environment data ecosystem that aggregates information from telecommunications companies, credit card providers, hotel systems, cruise terminals, and industry partners into three distinct functional spaces.
The Data Marketplace is the democratization layer. Over 50 shared datasets, accessible for free to registered users. Monthly hotel statistics. Visitor arrival forecasts. Cruise trends. A hawker stall owner in Chinatown can access the same market intelligence that was previously locked inside research departments at Marriott or Singapore Airlines.
The Data Sandbox is the collaboration layer. Here, STB partners mesh their proprietary data with national-level statistics to run predictive models. When travel patterns from Australia or China start shifting, the signals show up in the Sandbox weeks before they appear in the arrivals data.
The Private Space is the trust layer. Individual organizations, particularly SMEs, can upload company-specific data into a secure environment. They use Stan’s visualization tools to benchmark their performance against industry averages, without exposing a single number to competitors.
Three environments. One ecosystem. And it produces insights that have fundamentally changed how Singapore makes money from tourism.
The Day 3 Pivot: Where Data Becomes Revenue
Here’s the discovery that made me rethink what data readiness actually looks like in practice.
Through analysis of aggregated movement and transaction data, Stan identified that approximately 1 in 10 tourists in Singapore switches hotels during their trip. Not because of a problem. By choice. And the transition typically happens on Day 3.
That alone is interesting. What Stan revealed next was transformational.
Japanese and South Korean travelers were significantly more likely to begin their stay in mid-tier properties, then move to luxury hotels or integrated resorts like Marina Bay Sands for the final leg of their trip. Start practical. Finish premium.
Think about what this means. A Japanese traveler checking into a 3-star hotel on Day 1 is not a budget traveler. They’re a luxury traveler on a delayed timeline. By Day 3, they’re statistically ready to upgrade. The traditional approach, treating them as a mid-market segment for their entire stay, leaves enormous revenue on the table.
Singapore’s response was surgical. Targeted mid-trip upgrade promotions delivered via mobile apps and partner platforms. Digital nudges on Day 2, precisely when the data showed the decision window opens. Premium suite offers. High-end dining packages. Exclusive entertainment access. Not spam. Precision.
This specific intervention is credited with pushing the average per-visitor spend to S$1,804. Not by getting more tourists. By understanding the tourists already there.
Data is a commodity. Context is a moat.
Stan didn’t just collect data. It built what I call the Context Graph: the accumulated, connected record of decisions, behaviors, and outcomes that turns raw numbers into institutional judgment. The Day 3 Pivot wasn’t visible in any single dataset. It only emerged when movement data, transaction data, and accommodation data were connected, governed, and analyzed as a system.
Why This Matters Beyond Singapore
I’ve spent 25 years watching organizations invest in technology without investing in readiness. At HSBC, I led a transformation that reached 33,000 employees across the region, achieved 93% awareness and 72% participation rates. The lesson from that experience, the one I keep returning to: the technology was never the hard part. The organization was.
MIT’s Project NANDA research, published in July 2025, confirmed this at global scale. Across 300+ AI implementations, 95% of organizations are getting zero return from their AI investments. The researchers were blunt: “This divide does not seem to be driven by model quality or regulation, but seems to be determined by approach.”
Singapore’s tourism sector sits squarely in the 5%. Not because it has better AI than everyone else. Because it was ready for AI before it deployed AI.
When I map Singapore’s approach against the six dimensions of AI readiness we use at AIR APAC, the alignment is almost uncomfortable in its precision. Every dimension is addressed. Not accidentally. Architecturally.
Dimension 1: Leadership & Vision (22% weight)
Leadership carries the heaviest weight in our framework for a reason. It’s where most failures originate.
I know this personally. At HSBC, I watched a country CEO walk out of a transformation presentation after three minutes. No explanation. The problem wasn’t the initiative. It was the framing. We’d positioned it in a way that made him think “this is an HR thing.” We lost him in 60 seconds.
STB’s leadership didn’t make this mistake. They didn’t position Stan as “an IT project” or “a data initiative.” They positioned it as yield engineering infrastructure. The distinction matters enormously. “IT project” gets delegated. “Yield engineering” gets the CEO’s attention, because it directly connects to the number every tourism leader is measured on: revenue.
The July 2025 OpenAI partnership tells you everything about STB’s leadership posture. They didn’t wait for a vendor to pitch them. They became the first national tourism organization in Asia to formally adopt OpenAI’s technology. That’s not a procurement decision. That’s a strategic positioning decision made by leaders who understood what AI could do for their specific context.
The question for your city: Do your leaders understand AI deeply enough to lead the conversation, or are they waiting for a vendor to lead it for them?
Dimension 2: Data Readiness (20% weight)
This is where Singapore’s advantage becomes structural. And it’s the dimension that separates cities that use AI from cities that are ready for AI.
The Day 3 Pivot insight was only possible because Singapore had spent years building data infrastructure before it tried to extract intelligence from it. Stan was operational long before the OpenAI partnership. The data was already clean. Already governed. Already interoperable across agencies.
When the OpenAI MOU was signed, STB didn’t need to spend months cleaning data or negotiating data-sharing agreements between departments. The foundation existed. The AI had something meaningful to work with from day one.
Most cities I encounter have the opposite problem. I’ve seen organizations where a project that should take two weeks takes nine months, not because of technical complexity, but because data is hoarded across departments and nobody has the authority or the infrastructure to connect it.
The question for your city: If you signed an AI partnership tomorrow, would your data be ready? Or would you spend the first year just cleaning it?
Dimension 3: Skills & Capability (18% weight)
Stan’s three-environment architecture reveals something subtle about Singapore’s approach to capability building.
By giving SMEs access to the Data Marketplace and Private Space, STB didn’t just share data. It trained an entire ecosystem to think in data. A boutique hotel operator who starts by checking occupancy benchmarks in the Marketplace eventually learns to upload their own data into the Private Space. Then they begin running comparisons. Then they start making decisions based on patterns rather than instinct.
This is the shift from using AI tools to judging AI outputs. I call it the Auditor Mindset, and it doesn’t develop by accident. It develops when people have safe, structured environments to practice.
I learned the cost of skipping this step at an enterprise AI company in Houston. I was the CMO who was “too busy with marketing” to learn the AI product deeply. The CEO was blunt about it in front of the entire executive team. He wasn’t entirely wrong. I was using tools without understanding them well enough to evaluate their outputs. That moment reshaped how I think about capability: it’s not about knowing how to use AI. It’s about knowing when AI is wrong.
Singapore built this capability at the ecosystem level. The T-LEAP program trains tourism leadership talent specifically for an AI-driven future. The TIH platform, with its 100+ APIs, ensures that 1,800 registered companies can integrate data-driven tools into their own operations without needing a data science team.
The question for your city: Are your tourism operators trained to interrogate AI recommendations, or just accept them?
Dimension 4: Process Maturity (15% weight)
The Day 3 Pivot only works because Singapore redesigned its processes to allow real-time intervention.
Think about what the mid-trip upgrade promotion actually requires. It’s not just a marketing campaign. It’s a workflow that connects real-time guest location data to predictive behavioral models to targeted digital delivery to hotel inventory systems to payment processing. All within a 24-hour decision window on Day 2 of a visitor’s stay.
This workflow didn’t exist before Stan. It couldn’t have. The old process was: collect survey data quarterly, publish a report, hope hotels read it, wait for next quarter. By the time insights reached operators, the visitors were home.
AI accelerates whatever’s already there. If your process is “quarterly report, hope someone acts on it,” AI will generate that quarterly report faster. Singapore redesigned the process first. Made it real-time. Made it interventional. Then deployed the technology.
The Conrad Centennial Singapore offers a micro-example: AI-powered scheduling software reduced administrative time from two hours to 15 minutes. But the value wasn’t in the time saved. It was in what staff did with the recovered time: high-touch, high-value guest interactions that drive the per-day yield numbers.
The question for your city: Are your visitor management workflows designed for real-time intervention, or quarterly reporting?
Dimension 5: Governance & Ethics (15% weight)
When STB signed the OpenAI MOU, one of the three stated pillars was responsible adoption. Not innovation. Not speed. Responsible adoption.
This matters because tourism data is inherently personal. Movement patterns. Spending behavior. Accommodation choices. Dining preferences. Connected and analyzed at scale, this data can reveal intimate details about individual travelers.
Stan’s three-environment architecture is itself a governance structure. The Private Space exists because STB understood that businesses would never share data if they feared competitors could access it. The Data Sandbox requires partnership agreements. The Marketplace publishes aggregated, anonymized data only. Every layer has clear rules about what can be shared, with whom, and under what conditions.
Governance is not the Department of No. It is the Department of How. Singapore showed that governance, designed well, doesn’t slow down innovation. It enables trust. And trust is what convinces 1,800 companies to participate in your data ecosystem.
The question for your city: Do you have clear policies for how visitor data is collected, shared, and protected? Would your tourism operators trust your governance enough to share their data?
Dimension 6: Culture & Change Capacity (10% weight)
Culture carries the lowest weight in our framework. Not because it’s unimportant, but because leadership and data failures kill initiatives before culture has a chance to matter.
But Singapore has built something distinctive: a culture of participation across the entire tourism ecosystem.
Stan’s free Data Marketplace isn’t just generosity. It’s a cultural strategy. When a hawker stall owner can access the same visitor trend data as the Ritz-Carlton, you’ve democratized not just information but ambition. The 73% of hotel rooms with sustainability certifications, far exceeding the 60% target, reflects an industry that has internalized the direction, not just complied with it.
The 300,641 attendees at the 2025 Formula 1 Grand Prix. The 2 million cruise passengers. The record MICE rankings. These aren’t just events. They’re expressions of a city where government, industry, and operators move in the same direction because they share the same data, the same platforms, and the same understanding of what “success” means.
The question for your city: Does your tourism ecosystem move together, or does each stakeholder optimize for themselves?
The Uncomfortable Implication for 145 Cities
I work with tourism authorities across Asia-Pacific. I see what most cities are doing with AI: buying platforms. Installing chatbots. Building dashboards. The technology is often excellent.
And 95% of it, based on MIT’s research, will generate zero return.
Singapore’s lesson is not “build a Stan.” You can’t copy a platform. The lesson is: Singapore was ready for Stan before Stan existed. The leadership understood yield engineering. The data was being governed and connected. The workforce was learning to think in data. The processes were being redesigned. The governance frameworks were in place. The culture supported experimentation.
When the AI arrived, it had an organization to land in.
I tell a story in my work about the Context Tax: the hidden cost of deploying AI that doesn’t understand local context. Amazon learned this when Alexa launched in Singapore and couldn’t understand Singlish. Cities learn it when they deploy visitor management AI trained on European or American travel patterns and wonder why it fails in Bangkok or Kuala Lumpur or Manila.
Singapore avoided the Context Tax because it built the Context Graph first. It accumulated institutional judgment, digitized it, governed it, and made it accessible. The AI didn’t need to guess about Singaporean tourism patterns. The patterns were already mapped.
Where to Start
For the 145 member cities in the TPO network, and for every tourism authority reading this, the path forward is not a technology purchase. It’s a readiness diagnostic.
Assess honestly. Not where you want to be. Where you are. Across all six dimensions. A structured assessment that gives your city a score, identifies specific gaps, and prioritizes what to fix first.
Start with Leadership and Data. They carry 42% of the total weight for a reason. If your leaders can’t articulate an AI vision beyond “we need to go digital,” and if your data sits in disconnected silos across five departments, nothing else you build will hold.
Design for yield, not volume. Singapore proved that S$23.9 billion can come from 16.9 million visitors. The Day 3 Pivot didn’t require more tourists. It required understanding the tourists already there.
Build the Auditor Mindset. Train your tourism operators to evaluate AI recommendations, not just accept them. When an AI system recommends a pricing strategy or a marketing campaign or a visitor routing decision, someone in your organization needs to be capable of asking: is this right for our context?
The AIR APAC AI Readiness Assessment is designed specifically for this. Six dimensions. Weighted by actual impact. Calibrated for APAC cities. Not a 50-page strategy deck. An honest diagnostic and a clear path forward.
MIT’s research established consensus: the strategic positioning window is closing between mid-2026 and early 2027. Cities that assess and act now can implement in 90 days. Those that wait will find the 5% have already moved on.
Singapore didn’t wait.
The technology is ready. Is your city?
If you lead a tourism authority and want to understand what readiness looks like for your specific context, reach out. DM me or visit airapac.org. No pitch. Just a conversation about where you are and what the six dimensions mean for your city.