ChatGPT Optimization

This skill tunes an AEO program for ChatGPT specifically — usually the highest-priority engine given its reach. The cross-engine fundamentals apply; this skill says where to add weight. Behavior evolves, so treat these as current emphases and confirm by testing. It works for any business, B2B or B2C.

How ChatGPT builds an answer

ChatGPT answers two ways, and they call for different work:

• From trained knowledge (a snapshot with a cutoff): here it can only mention what the model already "knows" about a brand, shaped by how the brand appears across the broad web it learned from. This is won through entity strength and consistent presence over time.
• From live web search (when it browses): it retrieves from a broad web index and cites sources in real time. This is won through reachability, relevance, and freshness like other retrieval engines.

Across both, ChatGPT leans on established, encyclopedic sources (Wikipedia especially), brand mentions across the web, and author authority. So a complete program covers both the "be a known entity" game and the "be retrievable and quotable" game.

When to use this skill

• A brand cares specifically about ChatGPT visibility.
• Diagnosing why competitors appear in ChatGPT answers and the brand does not.
• Prioritizing AEO for a ChatGPT-heavy audience.

Before you start

Read aeo-foundation.md for the prompt set, competitors, and entity facts, plus a ChatGPT baseline if available. If it is missing, point to the aeo-foundation skill.

Where ChatGPT puts weight

• Entity strength and encyclopedic presence. It leans on Wikipedia and well-established references; a recognized entity with a Wikidata/Wikipedia footprint and consistent facts is named confidently. Prioritize entity-optimization.
• Brand mentions and corroboration. Consistent mentions across many trusted sources are a strong selection signal — invest in digital-pr and community-and-reviews.
• Author authority. Named, credentialed authors and clear E-E-A-T support citation — see author-authority.
• Web-search presence. Its search retrieves from a broad index; solid organic visibility and clean reachability feed it (confirm ai-crawler-access for OAI-SearchBot and ChatGPT-User, and ai-retrievability).
• Extractable, well-sourced answers. Clear answer-first content with citations is easy to quote — see answer-engineering and evidence-and-citations.
• Freshness for time-sensitive prompts, balanced with authority for evergreen ones.

Quick wins

• Establish or complete the entity: a referenced Wikidata item, Organization schema with sameAs, and identical core facts everywhere.
• Pursue third-party mentions in the sources ChatGPT tends to trust for your category.
• Add named, credentialed author bylines to key content.
• Confirm OAI-SearchBot / ChatGPT-User aren't blocked (run the ai-crawler-access checker).
• Add answer-first, cited passages to the pages tied to your top prompts.

Workflow

1. Baseline in ChatGPT (and its search mode where available): citations, sources, and who wins; note what it says from memory vs. when browsing.
2. Strengthen the entity — Wikidata/Wikipedia footprint, Organization schema, consistent facts.
3. Build mentions and corroboration across the sources it trusts for the category.
4. Signal author authority on target content.
5. Confirm web reachability for the search crawlers.
6. Re-test and iterate (cadence: see ai-visibility-tracking).

Output format

• A ChatGPT baseline and gap list (memory vs. browsing where distinguishable).
• Prioritized moves weighted to entity strength, brand mentions, and author authority.
• Target entities, sources, and pages to act on.
• (assumption) tags where behavior was inferred rather than tested.

Common pitfalls

• Weak or absent entity presence, so ChatGPT cannot confidently name the brand from memory.
• Relying only on owned content with little third-party corroboration.
• Anonymous content with no author authority.
• Blocking the search/user crawlers (or letting a CDN do it), so browsing can't reach you.
• Assuming ChatGPT weights sources like Perplexity — it leans more encyclopedic and less on raw recency/community.

Related Skills

• entity-optimization — the encyclopedic/entity presence ChatGPT favors.
• digital-pr / community-and-reviews — the brand mentions and corroboration.
• author-authority — the E-E-A-T signals it rewards.
• answer-engineering / evidence-and-citations — extractable, cited answers when it browses.
• ai-crawler-access / ai-retrievability — reachability for its search index.