You Don’t Need To Win the AI Race To Win in Travel

Written by: Dusan M.

Let's talk about AI strategy in travel. Specifically for companies trying to figure out where they fit. The non-chosen ones. Nobody tells you that the AI race was never yours to run. And that's actually good news.

The narrative right now is that every travel company needs to go all-in on AI or get left behind. That's not strategy. That's panic. That's how business dies. And trust me - ChatGPT will not make your business work if it's not working without it. It will only write the most bs possible to persuade you it will work.

Here's a more practical way to think about it.

Understand who owns what's needed.

The incumbents - the GDSs, the major OTAs, the large TMCs, large providers - have two things that no amount of money can buy and replace quickly: decades of historical data and deep domain expertise. They know the fulfillment chain. They know how a booking breaks, why it breaks, and what it costs when it does. They have the data to train AI systems that actually work in our industry. The ones who replaced their expert talent with resume logos will fumble, and by the time they figure out why, they will be out of the picture. Just look at Tripadvisor's issues. Pressure, pressure, pressure - deserved or undeserved - is creating a literal stagnation.

Let the Big Boys solve the big, complex, industry-wide problems. That's their job and they have the resources to do it. Your job is different and actually much better.

Focus on your customer, not the race.

If you're a tour operator, a TMC, OTA, TA or a travel tech provider - your advantage isn't data or resources. It's proximity to your customer and your expertise. You know your niche. You know what your customer needs before they ask for it.

That's where AI creates real value for you. Not by solving industry-wide distribution, marketing, or operational problems. By making the decision to use your service the best decision they make all day because you will take care of them no matter what. It's a service business, so - service.

Use AI to boost what you already do well. Not to replace your expertise - to amplify it. The travel company who uses AI to improve customer's decisions and relieve the pain of countless choices will outperform the one who built a generic chatbot because everyone else did.

Be pragmatic about data.

If you don't have large-scale data yet - that's fine. But start collecting it now.

Every customer interaction, every booking pattern, every service failure and recovery is data. Collect it intentionally. Structure it. Because the companies carving out their niche in the AI-powered future of travel aren't the ones with the biggest budgets today. They're the ones who understood early that their own data is their most valuable long-term asset.

You're not collecting data to compete with Sabre Corporation , Amadeus , Expedia Group or Booking.com. So don't. You're collecting it to own your corner of the market in a way no one else can replicate. Try to add to your data in any way possible. Data + your expertise = functional + feasible AI.

The practical filter.

Peter Drucker once said: "the purpose of a business is to create a customer."

So before any AI investment ask two questions: will it create more customers for my business, and does this make my customer's experience measurably better?

Not - does this make us look innovative. Not - are our competitors doing this. If the answer to both questions is yes, invest gradually. If the answer is maybe, experiment with a small budget and a clear success metric. If the answer is still maybe or no, ignore it regardless of the hype.

If you are wondering where to invest money with AI - in the only two things that matter. Data and expertise. There is some data available for free in our industry that could support small exploratory projects. There are some data providers as well. But keep in mind none of that data is as valuable to you as yours.

And please don't think that you will outsmart some of the smartest people in the world with vibe coding. I won't get into that. Just don't, please. Get an engineer that knows how to translate your expertise into a functional stack and how to turn every experience into data which feeds and trains the model. You will pay far less than you will to repair the damage if done haphazardly. Don't worry about the infrastructure yet.

For the Innovators, the Elon Musks and the Jeff Bezoses of Travel

LLM will not replace the domain expertise. AI eats only two types of food: domain expertise and data. That's the only bait that attracts the game.

We work in a cash- and labor-intensive service industry. You will not make a trillion-dollar empire building a new Tesla or reselling books. However you will do well if you excel at providing a great service. Because in our industry service IS the product. If you have a great idea, pitch it. Allow someone who walked your shoes to try to break it for you, so you can fix whatever you can before you go to the market with it. It's a margin-thin industry, so stay focused, persistent and use rejection as feedback. Every rejection is just another step closer to your vision especially if it results with you becoming great at what you do. Even Big Boys are receptive of great ideas. Pitch.

The bottom line.

The AI race in travel will be won by companies that know who they are, what data they have, what they know how to do well, and what customer problem they're uniquely positioned to solve. Not by companies that panic-bought AI tools because LinkedIn crowd told them to or Silicon Valley guy talked up on TED.

Be smart. Be pragmatic. Create a customer. Serve your customer. Collect data. Amplify your expertise. You will be surprised that along the way you might just end up transforming not just your business - but potentially the industry. Best service products were rarely the consequence of precise specs. They were mostly the result of a long grind focused on the customer. And that's what AI should help us do better.