The problem is not a lack of automation. It is a lack of teams that understand the business.
As artificial intelligence reshapes the economics of customer service, the travel industry is discovering that complexity resists automation in ways that simpler sectors do not. A decade and a half of operational data from co-sourcing firm 247 Travel Partner Desk reveals that human specialists achieve service quality 21 percent higher than AI-only systems — a gap that widens precisely when journeys unravel and passengers need someone who can think, not just respond. The deeper question emerging from this moment is not whether machines will replace people, but what kind of human judgment remains irreplaceable when every competitor holds the same technological tools.
- The race to automate travel customer service has hit a measurable ceiling: AI-only systems top out at 69% compliance, while human specialists surpass 90% in more than half of all operations.
- The real pressure points — midnight cancellations, medical emergencies mid-journey, multi-vendor rebooking — are precisely the scenarios where chatbots falter and brand reputation is won or lost.
- A hidden crisis compounds the problem: Latin American contact centers lose 30–40% of their agents annually, eroding institutional knowledge and producing the inconsistent service that no AI patch can fix.
- 247 TPD's answer is specialized co-sourcing — external teams embedded in client culture and KPIs — cutting operational costs 40–70% while holding quality steady through sector expertise and a 95% reduction in turnover.
- The argument lands at ConX 2026 in Palma de Mallorca, where over 1,000 travel executives will confront the conference's own thesis: in a world where AI makes everyone average, difference is the only strategy left.
The travel industry is living a familiar tension: artificial intelligence promises to slash costs and scale operations, yet some problems still demand a human being who genuinely understands the business. Operational data from 247 Travel Partner Desk, a co-sourcing firm serving cruise lines, airlines, and tour operators, puts concrete numbers to that debate. Human agents specializing in travel support deliver service quality 21 percent higher than AI-only systems. More than half of human-managed operations achieve compliance above 90 percent — sometimes reaching 100 percent — while AI alone maxes out at 69 percent, and AI-assisted hybrid models reach only 79 percent.
The gap exists because travel is not simple. A flight cancellation at midnight, a passenger needing to rebook across three airlines and two hotels, a refund dispute involving multiple vendors — these are not structured queries. Founder Gerardo Ariño frames the stakes clearly: when every competitor accesses the same AI tools, the difference comes down to the team behind those tools. Automation handles volume. Expertise handles complexity.
The business model built around this insight is specialized co-sourcing: external teams operating under the same KPIs, procedures, and brand culture as the client. The financial case is compelling — costs fall 40 to 70 percent depending on the market — but savings alone do not explain why Norwegian Cruise Line Holdings, Air Europa, and Viajes El Corte Inglés have chosen this path. The real advantage is sector specialization combined with stability. Contact centers across Latin America typically lose 30 to 40 percent of their agents each year, bleeding institutional knowledge with every departure. 247 TPD has reduced its own agent turnover by 95 percent, drawing on Gallup methodology and certified coaching across a 600-person, six-language operation based in Cali, Colombia, built over 15 years.
The timing of these findings is deliberate. The company is presenting them at ConX 2026, a travel distribution conference gathering more than 1,000 executives in Palma de Mallorca on June 2, under the theme: "In a world where AI makes everyone average, difference is your strategy." If AI is becoming a commodity, competitive advantage must come from elsewhere. Ariño's conclusion is direct: the problem in travel is not a lack of automation — it is a lack of teams that truly understand the business. AI solves one part. Trained, specialized talent solves the rest.
The travel industry is caught in a familiar tension: the promise of artificial intelligence to cut costs and scale operations, versus the stubborn reality that some problems still require a human being who understands the business. A new study from 247 Travel Partner Desk, a co-sourcing firm that handles customer service for cruise lines, airlines, and tour operators, offers concrete numbers to that debate. Human agents specializing in travel support deliver service quality 21 percent higher than AI-only systems, according to the company's operational data. More than half of the operations managed by human teams achieve compliance rates above 90 percent, with some reaching 100 percent. AI systems alone max out at 69 percent compliance. Even when humans assist AI—what the industry calls hybrid or AI-assisted models—the rate climbs only to 79 percent.
These numbers matter because the travel business is not simple. A flight cancellation at midnight in Barcelona is not a structured query. A passenger with a medical emergency who needs to rebook across three airlines and two hotels is not a chatbot scenario. Disputes over refunds, coordination between hotels and ground operators, last-minute itinerary changes—these are the cases that fill the real work of travel customer service. Gerardo Ariño, who founded and leads 247 TPD, frames the problem this way: when every competitor has access to the same AI tools, the difference comes down to who has the better team behind those tools. The question is no longer whether to use artificial intelligence. It is what distinguishes companies that know how to integrate it into their actual operations, their culture, their decision-making.
The broader context is that AI adoption in customer service is accelerating. Gartner reports that 38 percent of organizations have already deployed AI chatbots for customer support, up 25 percent in just two years. This wave has successfully automated the transactional layer—simple questions, standard bookings, basic troubleshooting. But it has not eliminated the need for judgment. Ariño notes that incidents at destinations, airline disruptions, post-sale complaints, and coordination across multiple vendors all require intervention from someone who can think, decide, and represent the brand authentically. Automation handles the volume. Expertise handles the complexity.
The business model that 247 TPD has built around this insight is called specialized co-sourcing: external teams that operate under the same key performance indicators, procedures, and brand culture as the client company itself. The financial case is strong. According to the firm's data, this model reduces customer service and continuous operations costs between 40 and 50 percent, and can exceed 70 percent in markets with high salary structures. But cost savings alone do not explain why major travel companies—Norwegian Cruise Line Holdings, Air Europa, Viajes El Corte Inglés, Avoris, and others—have chosen this path. The real advantage lies in the combination of sector specialization, low employee turnover, and the ability to scale quickly without sacrificing quality.
Turnover is the hidden metric that matters most. Contact centers across Latin America typically see annual turnover between 30 and 40 percent, according to industry research from Metrigy and SymTrain. That churn means constant retraining, lost institutional knowledge, and inconsistent service. 247 TPD has engineered its operation to reduce agent turnover by 95 percent, using Gallup methodology and certified coaches. The company operates from Cali, Colombia, with more than 600 professionals working in six languages. They have been doing this for over 15 years. That depth of experience and stability is not something you can buy off the shelf or replicate with a software update.
The timing of this argument is deliberate. 247 TPD is presenting these findings at ConX 2026, a travel distribution conference organized by Travelgate that will gather more than 1,000 travel industry executives in Palma de Mallorca on June 2. The conference theme—"In a world where AI makes everyone average, difference is your strategy"—captures the underlying anxiety. If artificial intelligence is becoming a commodity, if every player can access the same models and platforms, then competitive advantage must come from somewhere else. The company's thesis is that it comes from people: from teams that understand the travel business deeply, that can navigate its complexity, that can make decisions when the rulebook runs out. Ariño puts it plainly: the problem in travel is not a lack of automation. It is a lack of teams that truly understand the business. AI solves one part. Trained, specialized talent solves the rest.
Notable Quotes
When all competitors use the same AI tools, the difference comes down to who has the better team behind those tools.— Gerardo Ariño, CEO and founder of 247 Travel Partner Desk
AI solves one part. Trained, specialized talent solves the rest.— Gerardo Ariño, CEO and founder of 247 Travel Partner Desk
The Hearth Conversation Another angle on the story
So the headline is that humans beat AI by 21 percent. But what does that 21 percent actually mean in practice? What's the difference between a 69 percent AI resolution and a 90 percent human one?
The difference is whether the customer's problem actually gets solved. A 69 percent compliance rate means roughly one in three interactions doesn't meet the standard. In travel, that might mean a passenger with a flight cancellation gets a generic response instead of a rebooking across multiple airlines. A human agent with sector knowledge sees the full picture—the passenger's original itinerary, the available alternatives, the cost implications—and makes a decision that works.
But AI is supposed to get better. Why can't it learn these patterns?
It can learn patterns, but travel isn't just patterns. It's context. A flight cancellation in summer looks different from one in winter. A family with young children has different needs than a business traveler. An AI system can flag these variables, but deciding what to do requires judgment—the ability to weigh competing interests and represent the brand's values while solving the problem.
That sounds expensive. How does 247 TPD keep costs down if they're relying on human judgment instead of automation?
They've built a different kind of efficiency. Instead of trying to automate everything, they specialize. Their teams work only in travel. They stay in their jobs—95 percent lower turnover than the industry average. They use the same procedures and KPIs as the client company, so there's no translation layer. And they operate from markets where labor costs are lower without sacrificing quality. The cost savings come from stability and focus, not from replacing people with machines.
So the real argument isn't humans versus AI. It's about what kind of work actually needs to be automated.
Exactly. Automate the structured queries, the simple bookings, the routine confirmations. But keep humans in the loop for anything that requires context, judgment, or a decision that affects the customer's experience. The 21 percent difference is what you get when you stop trying to automate your way out of complexity and instead invest in people who can navigate it.
And that works at scale? They're not just a small boutique operation?
They operate 600 professionals across six languages from one location. They've been doing this for 15 years. They support major cruise lines, airlines, tour operators. The model scales because it's built on specialization and retention, not on trying to do everything with fewer people.