You Can't Vibe Code a Tour Operator

From the Travel Tech Podcast newsletter for episode 20, You Can't Vibe Code a Tour Operator.

Most AI pitches in travel come from people who have never had to file a rooming list. Alex Ragin has. He has been building software for tour operators since 2015, and when he talks about where AI actually works and where it quietly fails in production, it is coming from a decade of real client relationships, not a demo environment. For those of you following along, I wanted that perspective on the show.

He started as a two-person consultancy, grew an agency to 35 people by niching entirely into travel, and then productized seven custom builds into a single platform now used by multi-day tour operators booking trips that average $3,700 per person (some as high as $30,000). He has built itinerary tools, chatbots, and booking engines, and he has a clear-eyed view of which ones actually hold up in production. You can listen here: Apple Podcasts, also streaming on Spotify and YouTube.

The conversation kept returning to the same tension: the gap between what AI can do in a controlled environment and what breaks the moment real data, real suppliers, and real travelers are involved.

The gap between a demo and a departure manifest

That tension has a structural cause. Tour operators, airlines, hotels, and cruise lines are not variations on the same business. They are distinct operational models with incompatible data standards, fragmented distribution, and deeply specific edge cases. A multi-day tour operator managing 25 travelers across a 10-day departure has to track payment installments, passport details, dietary restrictions, rooming preferences, supplier confirmations, and last-minute cancellations, simultaneously, at scale, where one missed step creates a customer service failure. The travel industry has been running workflows like this largely unchanged for 15 years, and there is no general-purpose platform that handles all of it well because the domain knowledge required took years to accumulate.

AI is adding pressure, but mostly at the top of the funnel. Tour operators are already reporting that leads from ChatGPT and Gemini referrals convert at significantly higher rates than traditional organic search traffic. As AI overviews reshape how travelers discover destinations and experiences, the companies with structured, AI-readable content are picking up share from those without it. Inside operations, the picture is more mixed: one client using a customer support chatbot on top of the Tourseta platform saw resolution rates well below industry benchmarks, at a cost that did not justify the output.

This week's guest is Alex Ragin, founder of Zoftify (a travel software agency) and Tourseta (a booking and operations platform for multi-day tour operators). We covered the real failure modes of AI itinerary builders, why vibe-coded apps collapse against production edge cases, where AI tooling is genuinely improving his team's output, and what distribution opportunities he sees going uncaptured across the industry.

Key Takeaways

• AI search referrals are already outperforming organic on conversion: Tour operators are seeing leads from ChatGPT and Gemini convert at rates well above traditional Google traffic, and this is where near-term AI disruption is most measurable in travel.
• Itinerary builders still require manual validation at every step: Missing supplier data (something as specific as whether a whale watching tour runs in the morning) causes hallucinations that directly damage the operator's relationship with the traveler.
• Code is roughly 20% of what makes a travel platform work: The remaining 80% is domain knowledge, process design, and edge-case handling, which is why vibe-coded alternatives consistently fail to match purpose-built systems in production.
• The most productive AI applications in the agency are internal: Development assistance, UI/UX auditing, and unstructured data analysis are delivering consistent gains; customer-facing AI features remain experimental rather than standard.
• Cruise and corporate travel niches are still significantly underserved: Cruise tech integration remains technically painful, and vertically focused corporate travel management (for remote teams, support organizations, or specific industries) is a segment the major TMCs cannot serve with the specificity it requires.