One Hotel, Seven Versions turns a public-source hotel case into a reusable way to find where room facts, conditions and confirmation paths change between the systems a traveler encounters.
Tamaga has released the research method behind One Hotel, Seven Versions as an open online edition and fixed downloadable PDF.
The release is not significant because Tamaga published a white paper.
It is useful because it makes a difficult hotel-information problem inspectable:
A fact can remain visible across the travel web while the condition that makes it trustworthy disappears.
The study traces 16 traveler-critical facts across seven representation roles in one public-source hotel case. It records where a fact is precise, partial, contradicted or not observable, and it separates the evidence from Tamaga’s interpretation.
The online research, methodology, evidence labels, figures, references, limitations, version history and corrections route are public. There is no email gate.
What Tamaga released
The release contains four things a hotel or implementation team can inspect and reuse:
- A seven-version representation map covering the owned website, direct booking engine, structured data, search and local surfaces, OTA and review platforms, destination and editorial sources, and AI synthesis.
- A 16-fact evidence register covering identity, room configuration, access, timing, policy and commercial conditions.
- A bounded classification method distinguishing precise or confirmed, partial or generic, contradicted or erroneous, and not stated or not observable evidence.
- A repair sequence: align the fact, prove the consequential claim, distribute coherent representations, connect operational truth, and retain human confirmation where judgment is still required.
This is not a scorecard for ranking hotels.
It is a way to ask whether the versions still describe the same stay where a guest needs certainty.
Why seven versions matter
A hotel is no longer represented by one page.
The guest may encounter an owned room page, a booking engine, JSON-LD, a map or search result, an OTA listing, a destination guide and an AI-generated answer before reaching the property.
Those representations should not contain identical wording. They have different purposes and different limits.
The risk begins when compression changes the decision.
For example:
Connecting rooms on request
can become:
Connecting rooms available.
The subject survived. The condition did not.
The same failure can affect step-free access, late arrival, parking, cancellation, meal service, room layout or any other fact whose practical meaning depends on a boundary.
That is the problem the method is designed to expose.
What the public-source case showed
The first edition uses Hôtel Mont-Blanc Chamonix as an illustrative case. The property was selected because it is not digitally absent: it has a substantial official site, room pages, a direct booking application, search and OTA representations, destination material and a historical AI synthesis available for review.
Tamaga compared sixteen facts across the seven representation roles.
The result was not a simple pattern of “present” and “missing.” Different surfaces preserved different fragments of the stay. Some facts were precise. Some were generic. Some could not be observed with the available public method. In the consequential cases, the harder problem was often recovering the condition around the fact, rather than discovering the hotel name or an amenity label.
That finding changes the useful question.
Instead of asking only:
Can a traveler or AI system find this hotel?
the method asks:
Can it recover the room, condition and next action accurately enough to support the decision?
The case is evidence of a mechanism, not an estimate of industry prevalence. One hotel cannot establish how often representation drift occurs across the sector.
A method a hotel can use before buying more software
The practical starting point is deliberately small.
Choose one promise that could materially change a stay:
- a room that must fit a particular party;
- connecting rooms;
- step-free access;
- arrival after a stated time;
- parking that must be reserved;
- a dietary accommodation;
- or a cancellation condition.
Then trace it through the representations a real guest can encounter.
For each version, record:
- what the surface actually says;
- which source supports the statement;
- whether the condition remains visible;
- whether the guest has a safe action or confirmation path;
- who can correct the representation when it changes.
This does not require every surface to say everything.
An omission may be legitimate when a representation cannot express a condition honestly. A contradiction, false guarantee or lost action path is a different problem.
The goal is not maximal duplication. It is accountable agreement.
From research method to Tamaga product work
The research has two direct product consequences.
The Seven-Version Snapshot applies the method as a narrow, human-reviewed service. A property supplies one consequential promise; Tamaga traces its public representations, records the evidence and returns a repair map. It is not a full technical audit, reputation score or automated monitoring product.
Channel Lens is the corresponding Tamaga Hospitality mechanism for comparing governed meaning with selected representations. The current Hotel Demo proves that comparison with owned, generated and explicitly fictional examples. It does not claim live acquisition from OTA or AI services.
The distinction matters:
- the research defines and demonstrates the evidence problem;
- the Snapshot applies the method to one real public promise;
- Channel Lens makes representation comparison inspectable inside the product architecture.
None of the three should manufacture certainty that the evidence does not support.
Publication boundaries
One Hotel, Seven Versions is a public-source forensic study and an illustrative case.
It is not:
- an audit commissioned or endorsed by Hôtel Mont-Blanc Chamonix;
- a rating of the hotel’s hospitality or commercial performance;
- a legal, accessibility or technical-compliance assessment;
- evidence that every hotel has the same representation problem;
- or proof that structured data or AI visibility causes commercial outcomes.
The review period and inaccessible surfaces are identified in the research. The proposed structured-data example reviewed by the paper is not presented as verified live hotel markup. The direct booking application was present, but its live contents were not fully inspectable through the text-only method.
Those limits are part of the result.
An unknown should remain unknown rather than become a convenient claim.
Read and use the method
- Read One Hotel, Seven Versions online
- Download the fixed PDF from the Research edition
- Request a Seven-Version Snapshot for one consequential promise
- See how Channel Lens compares governed and represented meaning
The useful outcome of this release is not another hospitality report to collect.
It is a repeatable question a hotel can apply to one real decision:
Where did the guest-facing meaning change, and who is responsible for repairing it?
Publication note
Published by Tamaga on 9 August 2026. Substantially revised on 12 August 2026 to describe the released method and its practical consequence more directly.
Corrections
No corrections have been issued.