Javier Cabrerizo, Chief Strategy, Transformation & AI Officer At HBX Group, On How Agentic AI Is Transforming Business Decision-Making

“Strategic clarity starts with being very clear about where you want to go, without becoming too attached to the route you take to get there.

“Markets move, customer expectations evolve, and technology advances at extraordinary speed. Leaders have to be willing to adapt. But if every market shift sends you back to reconsider the strategy itself, you lose direction. The real challenge is distinguishing between noise, a change that requires a tactical response, and a shift that genuinely challenges one of the assumptions behind your strategy.

“At HBX Group, we keep the strategic destination stable while allowing execution to move quickly underneath it. We focus on a small number of long-term priorities and continuously adapt how we execute against them as market conditions change.

“AI is a good example. We treat AI as a structural transformation of the business rather than a technology upgrade, organized around three fundamental shifts: how we distribute, how we make commercial decisions, and how we operate at scale. Those three directions have not changed.

“That is also why AI at HBX Group sits with Strategy and Transformation rather than with Technology. It reshapes fundamental business choices so it must be governed alongside the strategy, not adjacent to it.”

“Markets become more responsive when decisions become better and faster. The challenge is that people can only process a limited amount of information, while today's markets generate an almost unlimited number of signals.

“AI changes that equation. It allows us to analyze booking patterns, demand signals, customer behavior, and market conditions at a speed and scale that would have been impossible only a few years ago. More importantly, it helps identify patterns and opportunities that would otherwise remain invisible.

“What agentic AI changes is that we move beyond insight and recommendation into action. An AI agent is given an objective and operates autonomously toward it, understanding the context, deciding what to do next, and executing where it is approved to do so.

“And oversight does not disappear; it moves up a level. Leaders spend less time making individual decisions and more time defining objectives, setting guardrails, and determining where human judgment adds the greatest value.

“We also need to be honest about the benchmark. AI can make mistakes, but so can people and existing processes. What matters is understanding the baseline you are improving on and putting the right controls in place.

“At HBX, we test agents in real conditions before production, monitor outcomes and errors, and increasingly use AI to identify failure patterns and improve other AI systems in real time.”

“Revenue management is about making the best commercial decision you can with the information available at that moment. AI changes how much information you can use and how fast you can respond to it. In the past, those decisions relied on historical patterns and a limited set of variables. Today, you can incorporate far more signals, from customer behavior and demand trends to inventory and market conditions.

“In travel, one distinction matters. Bookings tell you what converted. Search and demand signals tell you where demand is forming and where supply may be missing. The second is where much of the opportunity sits.

“But the advantage does not come from having more data. The advantage comes from turning data into better decisions. That is why at HBX Group we are building a common intelligence capability across the company that helps us anticipate market shifts and identify where to act to optimize pricing, inventory, and risk dynamically.

“As travel becomes increasingly algorithmic, a competitive advantage will come from the ability to combine AI-driven intelligence with human expertise to make better and faster commercial decisions. The goal is not a better pricing decision. It is a better commercial decision.”

“The starting point is not whether building, buying, or partnering is inherently better. It is understanding where ownership creates a strategic advantage and where speed matters more than control.

“If a capability is genuinely differentiating, closely linked to your proprietary data, operating model, or customer relationships, you should build it. If speed is critical and acquiring gives you technology, talent, or market access that would take years to develop internally, then acquisition may be the right answer. And partnerships are often underestimated. The right partner can help you create value much faster than either company could alone.

“With AI, we use all three approaches. The capabilities closest to our proprietary data and how we operate; we build ourselves. Where proven internal capabilities can be extended into propositions alongside ecosystem partners, we co-create. And where a specialist can materially accelerate a capability or our route to market, we partner.

“What matters is understanding the trade-offs: speed, economics, integration, control, and long-term strategic importance. The answer is rarely permanent. A capability you partner for today may become one you want to own tomorrow.

“The mistake many companies make is treating these as technology decisions. They are strategic decisions about where future competitive advantage will come from.”

“The first question is always the simplest one: which problem is being solved?

“When excitement builds around a new technology, many companies start with the technology and then go looking for a problem. The businesses with the greatest long-term potential do the opposite. They start with a problem customers genuinely care about and use technology as the means to solve it.

“A compelling problem and an exciting technology are not enough on their own. Long-term winners combine customer adoption, scalable economics, and a credible ability to capture value.

“For AI specifically, the most important consideration is differentiation. As AI models become increasingly accessible, access to the model itself is unlikely to be a sustainable advantage. What is harder to replicate is proprietary data, specialized domain expertise, and the ability to apply both to a specific problem in a way that creates measurable value.

“One test I like is to mentally remove the technology from the pitch. If the technology disappeared tomorrow, would customers still care about the outcome being delivered? The strongest businesses are those where the answer remains yes.

“Ultimately, lasting companies are built on solving real customer problems better than anyone else. Technology may create the opportunity, but customer value is what sustains it.”