ID Travel Group holds direct relationships with roughly 350 hotels, plus airlines, with real-time availability and dynamic pricing passed in at agreed net rates. Nothing is committed and nothing perishes. What varies is not whether a booking happens, it is which property receives it, and that single choice decides the margin. Every system below exists to make that choice well, at the moment it is made.
An earlier version of this page assumed ID Travel bought blocks of rooms and carried the risk of selling them. That was wrong, and Maurice corrected it. This version is built on the real model.
ID Travel holds direct relationships with roughly 350 hotels plus airlines. Availability and pricing arrive live, at agreed net rates. There is no committed inventory and nothing expires unsold.
Three consequences drive everything that follows. Discounts differ by property, so two bookings of equal value to the client are not of equal value to you. Overrides pay against annual production goals, which makes them thresholds rather than rates. And some partners carry non-financial value that no commission column records. This is a steering problem, not a distribution problem.
The companion analysis of the GLO scorecard, scored against your own criteria without changing a value, is at /glo/.
Hex, the platform moving into the commission and data layer between luxury hotels and travel advisors, does not market artificial intelligence anywhere on its public site. Not on the homepage, not on the agency page, not on the hotel page. Their language is automation and reconciliation.
Hex is the intelligent layer connecting luxury hotels and travel advisors. Stop chasing data, and start owning your production, commissions, and partnerships through a single, unified system of record.MY-HEX.COM HOMEPAGE, THE CLOSEST THEY COME TO AN AI CLAIM
They may well use machine learning inside the product. Matching a hotel's booking record to an advisor's record is exactly the kind of problem it suits, and the app is login gated so there is no way to know from outside. The verifiable claim is narrower and more useful: nobody in this category is competing on it out loud yet.
But the sharper point does not depend on what is inside Hex at all. Whatever technology they have serves their product, which is reconciliation. None of it decides which of your 350 partners a booking should go to, and none of it makes an advisor reach for you first. Hex is moving to own the layer your distribution runs through, and a business whose entire channel is advisors cannot afford for someone else to own that relationship. The defense is not better plumbing. It is being the partner an advisor reaches for first.
ID Travel Group sells luxury travel through human advisors. In that market, technology that announces itself destroys value. A chat bubble popping up with "Hi! How can I help you today?" makes a luxury brand cheaper, not smarter.
The technology does the work nobody wants to do. The destination specialists stay the face of the business. The specialists end up with more time on the conversations that need a human, not less. Any system that fails this test does not get built.
This is also the honest answer to what the industry fears. The threat was never technology answering questions. It is competitors serving advisors faster than you can.
One advisor inquiry, followed from the moment it arrives to the moment it lands on a property. Five systems, each pointed at the same decision.
An override is a threshold, not a rate. A booking sent to a partner sitting just below its production goal is worth far more than its percentage suggests, and one sent to a partner that will not reach its tier this year is worth only the discount. This tracks every partner's position against its goal and surfaces it at the moment of booking.
Nobody can hold that across 350 partners in real time while also serving an advisor on the phone. The same booking, steered one property sideways, can be the difference between clearing a tier and missing it.
Your network is not interchangeable. Some advisors sell Africa, some sell Caribbean shoulder season, some move groups, some have gone quiet for six months. When a partner is close to its threshold, the question is never "email everyone," it is which fifteen advisors have actually sold that destination in that season before.
Blasting the whole list trains advisors to ignore you. Sending nine people something they already know how to sell does the opposite.
An advisor describes a client in one sentence. Two adults, anniversary, first week of March, high end. Back comes a formatted, client-ready proposal built from the partners that best fit the brief and sit closest to a threshold, with the negotiated perks attached.
This is the system that changes your position, because it does two jobs at once. The advisor gets something in front of their client in minutes instead of an hour, and every proposal is pointed at the partner where that booking is worth the most to you. Advisors route business to whoever makes them look good fastest.
An assistant trained on your own material. The partner list, negotiated perks, protection and insurance terms, group policy, and which specialist covers which region. It lives inside the advisor portal and answers the lookup questions that consume a specialist's morning and produce nothing.
It never quotes a rate it is unsure of, never confirms availability, and never pretends to be a person. When it does not know, it captures the question and routes it with the full thread attached.
The offers page is not marketing. It is your partner terms, facing the only people allowed to act on them. Every offer carries a real validity window, reflects the terms currently agreed, and comes down when they change.
An advisor who pitches an expired perk to a luxury client looks incompetent in front of that client, and remembers who caused it. That cost never shows up in a report.
This is also where the bonus commission properties should surface on their own, generated from the same data that drives the ranking rather than assembled by hand each time.
This is the steering desk as it would look inside your portal. The numbers below are illustrative, built to show the shape of the system rather than to report anything real. The companion analysis at /glo/ is the opposite: no simulation, only your own file.
Nothing here requires a decision about all five systems at once. Phase one is small on purpose, because it should prove itself before anything else gets funded.
The override analysis at /glo/ shows this cannot be skipped. Make partner position against goal visible, then point the network at whoever is within reach. It needs no change to how anyone works, and it pays for itself out of tiers that would otherwise be missed by a handful of bookings.
Built on what phase one proves. This is the piece that changes how advisors choose between you and the alternative, and it aims every draft at the partner where the booking is worth the most.
Real value, less visible, and far easier to justify once the first two are running and measured against a number.
These get measured before anything is built. Without a baseline there is no honest way to prove the work later, and a number nobody baselined is just a story.
Override tiers reached, and how many were missed by a small margin. Everything else on this page is a means to move that one number.
Blended take on equivalent bookings, before and after steering. The point is that the same volume earns more.
How many of the network booked at all this quarter, versus how many exist.
What share of volume comes from your top ten advisors. High concentration is a risk sitting quietly.
Advisor asks to human answers. Speed is what makes you the easy partner to use.
A visibly robotic assistant on a luxury site does real damage. It stays assistive, never performs personality, and hands to a human quickly.
A wrong perk reaching a client is worse than a slow answer. The system quotes only from verified inventory and escalates anything uncertain.
Advisors are creatures of habit. If it is not obviously faster than emailing a specialist they already like, they will not use it.
An earlier draft of this plan called the required IATA number a leak. Given the model it is not, it is the wall that keeps consumers out of trade rates, and it should stay. The narrower question is whether legitimate buyers who lack the number on hand, new advisors and contractors under host agencies, are being turned away with it.
Every system here assumes override positions and advisor booking history can be read from wherever they live now. This is no longer a hypothetical risk. The scorecard analysis at /glo/ found the override field populated on 37 of 544 properties, and every European partner missing it entirely. Connecting and completing that data is the first real task, and nothing above works without it.
A system that ranks by margin can quietly start sending people to the wrong hotel. Fit is evaluated before commercial terms, never after, and the ranking shows the specialist why a property placed where it did. One bad stay costs more than one override, and an advisor who feels steered stops trusting the desk.