April 22, 2026
Every year around January someone sends me a headline about Caribbean tourism being back. It never really left. The Caribbean Tourism Organization confirmed what I've been watching since I ran SXM Deals in 2012 - international stay-over arrivals rose 2.5% in 2025 to an estimated 35 million visits, with another 3% to 4% projected for 2026.
The demand was always there. What changed is the booking cost.
Not the cost to the guest. The cost to you - the operator trying to convert that demand into revenue while OTA commissions climb, AI search reshapes how guests find you, and your booking system either works or quietly bleeds your margin. That's what's different in 2026.
The CTO data has a detail that I keep coming back to. In 2025, arrivals from South America jumped 23.7% to 2.4 million visits. Meanwhile Canada dropped 5.3% and Europe fell 3.3%.
That's a real shift in who's showing up at your dock. The South American guest booking a snorkel trip in Philipsburg doesn't necessarily behave like the Canadian who's done your Christmas Winds catamaran run for years. Different languages, different comfort with credit cards online, different WhatsApp-versus-email instincts, different ideas about what a deposit even means.
I've spoken to operators who haven't updated their pre-trip confirmation emails in years. Those emails still assume everyone speaks fluent English and has a US billing address. That gap is costing bookings right now.
A lot of operators I speak to are worried about AI search eating their traffic. I'd say the worry is real but slightly misdirected. What AI agents actually do is answer questions like "best snorkeling tours in St. Maarten" and surface a direct booking link. If your site loads and your checkout works on a phone in under 90 seconds, you're probably fine. The ones who get hurt are running seven-step flows with no mobile optimization - the AI recommends someone else instead.
Mobile checkout is no longer a nice-to-have. It's the whole game. If a guest finds you through ChatGPT, Google AI overview, or a travel blog at 10pm and can't complete the booking in two taps, that booking goes to whoever can. I watched this pattern play out slowly from 2019 through 2022. In 2025 it accelerated. In 2026 it's just the baseline.
Funny story - not actually funny. An operator I know in Marigot told me his all-in cost per Viator booking runs 28% to 30% once you stack the commission, their payment processor, and the currency conversion hit. He still thinks it's worth it because he doesn't have a direct booking strategy.
I don't think it's worth it at that percentage. And the math is getting worse.
The OTAs are not bad. Viator and GetYourGuide put your product in front of people who would never have found you. But if 80% of your bookings flow through a channel taking 25%+, you're a subcontractor with a boat. The operators doing well in 2026 treated the OTAs as acquisition and built something to capture the repeat guest directly. The direct-channel resurgence is real, and it happened because operators finally made their own booking experience good enough to compete.
Caribbean hotel occupancy dipped to 63.7% in 2025, down from 65% in 2024, even while average daily rates rose. That's the market being busy and competitive at the same time. For tour operators, the equivalent is running full schedules and still getting squeezed - because the cost of converting a booking went up even if the volume of demand didn't go down.
No-shows haven't improved. Last-minute cancellations are still a factor. The guest who books three activities and cancels two at 7am because "it looked cloudy" has been showing up every season.
What's changed is that in 2026, the operators with automated deposit systems, pre-trip reminders, and self-serve reschedule links are eating the margin that used to evaporate. That gap is measurably wider than it was three years ago - I see it in the bookings that run through Junglebee every season.
CTO is projecting 5% to 7% more cruise visits in 2026, on top of the 3% to 4% stay-over growth. More people on islands is a real tailwind. But that tailwind doesn't fix a checkout flow that dies on mobile. It doesn't close the gap between your OTA revenue and your direct revenue. It doesn't answer the guest who found you through AI search at 11pm and wants to confirm before she goes to sleep.
The operators who are winning right now did something boring. They made their booking process work on a phone. They set a deposit that actually deters flaking. They stopped manually chasing balance-due payments. None of that is glamorous, and none of it shows up in a CTO report. But that's the real answer to what changed. The demand was always there. What changed was who was set up to capture it without bleeding margin in the process.
I've been watching Caribbean booking demand since 2012. The numbers go up and down. A blown-out wind day will cancel your most expensive charter no matter how well the rest of the month looked.
The operators who survive all of that run a tight operation. Deposit at the moment of commitment. Reminder automated. Reschedule handled in one click. Money landing in their local bank without a three-country detour.
If you're not there yet, check what it actually costs to get there. It's usually less than one no-show.
