Short answer: hotel mapping matches records from different suppliers to one master property, and room mapping matches their differently named room types to one canonical room. Without it, your site shows duplicate hotels, hides the cheapest price behind a second listing, and compares rooms that are not actually the same. It is solved with master hotel IDs, specialist mapping providers, or in-house matching, and usually a mix of all three.
Why the same hotel looks different in every feed
There is no universal hotel ID that every supplier uses. Each bed bank, aggregator and channel assigns its own property codes and maintains its own content: names, addresses, coordinates, star ratings, photos. The same beachfront property can arrive as "Grand Palm Resort and Spa" from one supplier and "Grand Palm Resort (ex Palm Beach Grand)" from another, at slightly different coordinates, with different star ratings.
The causes are mundane: rebrands and renames, transliteration of non-Latin names, chains listing towers or wings as separate properties, franchises changing flags, and suppliers sourcing content from each other with errors compounding along the way. GIATA and Vervotech, two specialist providers, both describe supplier IDs that change without notice as a routine operational hazard. If you take feeds from several sources - which is the normal setup for any serious hotel API integration - these differences land in your database on day one.
Why rooms are even harder than hotels
Hotels at least have addresses and coordinates to anchor a match. Rooms have only free text. One supplier's "Deluxe King Room" is another's "King Deluxe" and a third's "DLX-K", and none of them tells you reliably whether the view, board basis, bed setup or cancellation terms are the same. Suppliers also bundle attributes differently: one sells "Deluxe Room with Breakfast" as a distinct product, another sells "Deluxe Room" with breakfast as a rate-level option.
Room mapping therefore has to parse names and attributes, normalise them into a structure (room class, bed type, view, board, occupancy) and then decide what is genuinely the same product. This matters commercially: if you cannot say two offers are the same room, you cannot honestly show the cheaper one as the best price for it.
What bad mapping actually costs
- Duplicate listings. The same property appears two or three times in results, which looks broken and splits your click-through across entries.
- Lost best prices. If supplier B's cheaper rate sits on an unmapped duplicate, the customer sees supplier A's higher price on the "main" listing and may leave for a competitor showing the better rate.
- Lost margin. When you cannot compare suppliers per room, you cannot systematically buy from the cheapest source. The spread between two suppliers for the same room is exactly the margin bad mapping throws away.
- Booking errors and complaints. Matching a sea-view room to a standard room, or the wrong board basis, produces guests who did not get what they paid for.
- Wasted content work. Teams end up manually merging listings that software should have merged.
The three mapping approaches
| Approach | How it works | Trade-offs |
|---|---|---|
| Master IDs / multicodes | Adopt an external master identifier per property and store each supplier code against it, so every feed resolves to one record. | Clean and stable, but depends on the ID provider covering your suppliers and keeping pace with changes. |
| Specialist mapping providers | Send your supplier content to a mapping service that returns matched hotel and room pairs via API, maintained continuously. | Fast to adopt and accurate at scale; an ongoing subscription cost and one more dependency. |
| In-house fuzzy matching | Match on normalised name, address, coordinates and phone with similarity scoring; parse room names into attributes yourself. | No licence fees and full control, but real engineering effort, and edge cases (chains, renames, non-Latin names) never stop arriving. |
In practice these are combined: a master-ID backbone or mapping provider for hotels, plus rules and review queues for the long tail, plus attribute-based matching for rooms. The right mix depends on how many suppliers you run and how much engineering you want to own.
Named mapping providers
Three names dominate this niche, as surveyed in AltexSoft's comparison of hotel mapping tools:
- GIATA maintains Multicodes, a widely used master-ID system that assigns one identity per property and cross-references supplier codes against it.
- Vervotech offers AI-based hotel and room mapping APIs and publishes coverage across hundreds of supplier feeds.
- Gimmonix provides mapping as part of a wider hotel distribution technology stack.
Accuracy claims are the providers' own and worth validating against your actual supplier mix during a trial: run a sample of your inventory through the service and manually audit a few hundred matches in the destinations you care about before committing.
Duplicate suppression in search results
Mapping produces the data; duplicate suppression is what the customer sees. Once every supplier record resolves to a master property, your results page groups all offers under one listing and shows the best lead price, with the room-level comparison behind it. The search index stores master properties, not supplier records, and each cached price is tagged with its supplier so the booking flow knows where to buy.
How mapping fits into a real integration
In the builds we deliver, mapping is a pipeline stage, not an afterthought. Static content from each supplier is ingested on a schedule, resolved against the master property store, and flagged for review when confidence is low. Room offers are normalised into attributes at search time so the hotel booking engine can group and rank them, and unmapped records are quarantined rather than shown as duplicates. The same pipeline serves a consumer site, a B2B travel portal or a white label deployment.
If you are still choosing which feeds to connect in the first place, start with our guide to choosing a hotel API and the overview of what a bed bank is; the more suppliers you add, the more the mapping layer becomes the thing that decides whether extra feeds add value or just noise.
This article is general information about travel technology and online marketing. It is not legal, tax or financial advice, and advertising platform policies change often. Check the current policy documents and take professional advice for your own situation.