A booking platform that ends the commission: how an operator took its reservations back from the marketplaces.

The challenge

Renting the customer relationship, one booking at a time.

Every reservation arrived through a listing marketplace. The marketplace took a percentage of each booking, held the guest relationship, set the cancellation rules, and could change any of it without asking. The operator was paying a variable fee, forever, for demand they were increasingly generating themselves through their own reputation.

This is the arithmetic that makes the case, and it does not need anyone's private numbers to be true. A marketplace fee is a percentage of revenue with no end date. A platform is a fixed cost with no percentage. The two lines cross, and after they cross the gap widens every year the business keeps operating.

What kept the operator on the marketplace was not the demand. It was that leaving meant building a booking engine, taking payment, managing the stay, and answering guests, all of which the marketplace was doing.

What we built

The full stack the marketplace was providing, rebuilt as the operator's own property.

• Booking engine and checkout. Guests search availability, book, and pay on the operator's own site, with the reservation landing directly in their system.
• Operator portal. The team runs reservations, availability, and the day to day from an admin portal built around how they actually work.
• Guest portal. Guests get their own space for the stay, so the relationship continues after the booking instead of ending at checkout.
• Live local data. Conditions and local information that refresh on their own rather than going stale in a page nobody remembers to edit.

Front end and back end, bilingual in English and Spanish throughout, built as one system the operator owns outright rather than a template rented from another vendor. Payment and guest data stay in the operator's own stack.

The client has not released booking figures, so this study reports what shipped and what changed, without numbers we cannot show you the source for.

The results

• Booking engine and checkout
• Operator portal and guest portal
• Bilingual, English and Spanish
• Self refreshing local data

Most bookings now arrive direct rather than through the listing sites, so the fee on those reservations stays with the operator instead of the marketplace, and the guest relationship belongs to the business.

Where these numbers come from: No client figures are published here. The operator has not consented to release booking data, and we do not publish numbers we cannot attribute. The qualitative outcome is the operator's own description of the change.

Key insights

• The fee is not the problem. The percentage is. Operators tend to compare a marketplace fee against the cost of a build and conclude the build is expensive. That comparison is wrong on its face, because one of those numbers recurs as a share of everything you ever earn and the other happens once.
• Owning the booking means owning everything around it. You cannot take reservations back without also taking on payment, the stay, and guest questions. Half of this project was not the booking engine at all. Underestimating that is how direct booking projects stall at a nice looking site with no operational spine behind it.
• Leaving the marketplace is not the goal. The listings still bring demand from people who have never heard of the operator. The point is to stop paying a percentage on the guests who came looking for this operator specifically, which is the segment a marketplace adds nothing to.
• Bilingual is a structural decision, not a translation task. Retrofitting a second language into a booking flow after launch means touching every string, every email, and every date format. Deciding it up front costs a fraction of what it costs later.