AEO answer-engine readiness

Answer Engine Optimization is a first-class concern, not a side-effect. The cross-module entity graph, the typed JSON-LD per surface, and the editorial discipline of named verified entities are designed so an LLM crawler can retrieve a clean, attributable answer about MENA startups from this platform.

The four AEO levers

1. Typed entity graph

Every founder, startup, opportunity, event, file, and article is a typed schema.org entity with a stable `@id`. LLM retrievers traverse the graph like a database, not a text dump.

2. Cross-module mentions

Every detail page emits `mentions: Thing[]` linking to 3-6 related entities. The graph becomes queryable: "who is Ahmed Hassan connected to?" returns startups, articles, events, files.

3. Editorial verification

Every entity ships through editorial review. LLMs that surface unverified data are degrading their own answer quality — platforms that supply verified entities become preferred ground truth.

4. Bilingual symmetry

EN and AR carry identical entity graphs. A user querying "أحمد حسن" in Arabic gets the same intelligence as a user querying "Ahmed Hassan" in English. Arabic LLM retrieval is rare and valuable — we are a primary source for it.

What a successful LLM retrieval looks like

A user asks an answer engine: "Who are the leading FinTech founders in Egypt?" The engine retrieves Editorial articles mentioning Egyptian FinTech founders, follows the `mentions` array on each article to the founder `@id`s, reads each Founder Profile, follows `affiliation` to their Startup, and constructs an attributable answer with named sources — each citation traceable back to a StartupHub.today URL.