# Images & content A property page travels far. The same record reaches a traveller browsing on a metasearch engine, a B2B buyer reading on an aggregator, an operator looking at it in your console, and an Adragent reply summarizing it. **Images and content** are what make those views feel like the same property told well — and a chain operating in 5 languages keeps each translation in sync from one canonical record per property. ## What "content" covers | Layer | What it carries | | ---------------------------- | ------------------------------------------------------------------------------- | | **Identity content** | Property name, official identifiers, contact info — typically not multilingual. | | **Marketing content** | Property descriptions, hero copy, key selling points — multilingual. | | **Image content** | Property photos, room photos, captions, ordering, language-specific variants. | | **Amenity & feature labels** | Multilingual labels for hotel-level and room-level attributes. | | **Policies** | Cancellation, child, pet, smoking, payment-at-check-in policies — multilingual. | | **Special instructions** | Check-in / check-out details, accessibility notes, transport hints. | Every layer can be in multiple languages, with provenance — which value came from which source, which translation is yours, which is the original. ## Multilingual is the default Adrasis treats multilingual content as the norm: * A property's English description and Turkish description coexist on the canonical record, locale-tagged. * A property image can have an English caption and a Spanish caption against the same image asset. * An amenity carries labels in every language your organization operates in. Travellers see the locale they request; operators see the locale they prefer. The catalog does not duplicate properties by language — it carries language *on* the record. Adding a new language is a content task, not a structural one. ## Images — what the catalog records For each image: * **The asset** — the photograph itself. * **A caption** per language — what it shows, in the operator's words. * **Ordering** — first image, second, third; the gallery order travellers see. * **Tags** — which room it depicts, what feature it highlights, whether it is the property hero or a detail shot. * **Per-channel exposure** — some channels accept many images, some only a few; you decide which surface where. * **Provenance** — which source contributed the image (a content feed, a direct upload, a partner sync). You add, replace, reorder, and tag images through the property's content surface. Adragent can do bulk operations — *"Replace the hero image on every Antalya beach property with the new sunset shot."* ## Description — what good catalog content looks like A few practical heuristics: * **The first paragraph carries the weight.** Travellers and snippet generators read the first 150–200 characters first; if the property has one differentiator, it goes there. * **Specifics over adjectives.** *5-minute walk to two metro stations* beats *centrally located*. * **Multilingual means re-thought, not translated.** A Turkish description for a Turkish-market traveller is not the English one machine-translated; it speaks to the audience directly — different cultural references, different selling points. * **Re-read after every catalog change.** A new room type, a renovation, a category change — the description should reflect it. The catalog's content surface is alive; you revisit pages, refine descriptions, replace tired imagery without rebuilding from scratch. ## Per-language images Sometimes the right image varies by language: * A property's hero image in English-language markets might be the pool-and-bar shot. * The same property in a Turkish-language market might lead with a family-around-a-table image. * A Middle East market hero might emphasize the gardens and shaded outdoor spaces. The catalog allows you to set per-language image ordering — same canonical image library, language-specific gallery sequence. You do this where it matters; you do not have to do it for every property. ## Content from multiple sources If the same property arrives in your catalog from multiple sources (a channel manager, a content feed, a direct contract), the canonical record reconciles them: * Descriptions from one source merge with translations from another, with provenance. * Image libraries from multiple sources combine; duplicates are merged by content fingerprint. * Amenity labels from one source's vocabulary translate into your canonical taxonomy through mapping. * Operator overrides (your edits) win where you have set them. See [Canonical mapping](/console/mapping-and-curation/canonical-mapping.md) for the model. ## Per-channel content exposure Not every channel uses every content asset: * A high-fan-out metasearch feed may only display a hero, a short description, and a few amenities — even if your catalog carries dozens of images and a long description. * A B2B aggregator may consume the full content payload, including operational notes. * A direct-booking channel may carry your full marketing voice, including bespoke video and structured content blocks. The platform sends each channel what it asks for; what does not fit a channel is not stripped from the canonical record. ## Adragent and content Common phrases that map to Adragent content operations: * *"Find every property without a hero image and queue them for content review."* * *"Add a Turkish translation of the description on the Marina property."* * *"Replace the second image on every Antalya beach property with the renovated-pool shot."* * *"Show me properties where the description is identical across English and Turkish — they likely need a real Turkish version."* * *"Apply our new family-friendly cancellation policy template to every property tagged Family-friendly."* Read operations return immediately; writes preview before applying. ## Where to next * **The classifications that pair with content** → [Categories](/console/catalog/categories.md) * **The place taxonomy for the property's location** → [Location & geography](/console/catalog/location-geography.md) * **The AI Resolve operator action that fills in catalog detail** → [AI Resolve](/console/catalog/ai-resolve.md) * **The mapping system that reconciles content across sources** → [Canonical mapping](/console/mapping-and-curation/canonical-mapping.md) --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://adrasis.gitbook.io/console/catalog/images-and-content.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.