The HEO Implementation
Checklist

The Entity Audit is the mandatory first step of every HEO engagement. It establishes the current state of the entity across all three discovery layers — SEO, AEO, and GEO — before any optimization work begins. Skipping the audit means building without a baseline, which makes it impossible to measure progress, identify the highest-priority gaps, or sequence the work correctly. The audit has three components: an AI presence assessment, a structured data audit, and a technical SEO review. The AI presence assessment queries ChatGPT, Perplexity, Gemini, and Copilot with the entity's category, location, and primary service terms — documenting exactly how each system currently represents the entity, what it gets right, what it gets wrong, and whether it mentions the entity at all. The structured data audit examines every Schema.org markup element currently deployed on the site, checking for missing required properties, incorrect @id references, entity fragmentation across pages, and the absence of the entity types that AI systems prioritize. The technical SEO review checks crawl health, AI crawler permissions in robots.txt, canonical URL consistency, and the indexing status of the pages that carry the most important entity signals.

The SEO Foundation Build is the phase that makes the entity legible — not just to traditional search engines, but to every AI retrieval system that depends on indexed, crawlable content to populate its answers. This phase is not about keyword rankings. It is about infrastructure: the technical and structural signals that tell all discovery systems who the entity is, what it does, where it operates, and why it is authoritative. The most critical output of this phase is a complete, consistent, cross-platform entity identity — the same name, address, phone number, and canonical URL appearing identically in every location where the entity has a presence. Entity fragmentation — the condition in which the entity appears under multiple name variants, URL formats, or address representations — is one of the most common and most damaging conditions that HEO audits uncover. AI systems build their understanding of an entity by aggregating signals from multiple sources. Fragmented signals produce a fragmented understanding, which produces lower citation confidence. The SEO Foundation Build eliminates fragmentation and replaces it with a coherent, machine-readable entity identity.

The AEO Layer Build is the phase in which the entity transitions from being indexed to being answer-eligible — from appearing in search results to being cited in AI-generated responses. This transition requires three things that the SEO foundation alone does not provide: question architecture, entity attribution, and authority density. Question architecture is the practice of restructuring content around the natural language questions that users ask AI systems. AI answer engines do not retrieve content the way search engines rank it. They extract the most direct, declarative, well-attributed answer to a specific question from the available indexed content. Content that is not structured around explicit questions and direct answers is not answer-eligible, regardless of how well it ranks in traditional search. Entity attribution is the practice of ensuring that every piece of content is explicitly connected to a named, credentialed entity — that the author is identified, the organization is named, and the relationship between the author, the organization, and the topic is documented in both the content and the structured data. AI systems use entity attribution to assess the credibility of a cited source. Anonymous or weakly attributed content is less likely to be cited, even when it is technically accurate. Authority density is the cross-platform citation record that AI answer engines use to assess whether an entity is genuinely authoritative on a topic — not just self-declared, but corroborated by other sources.

The GEO Layer Build is the most advanced phase of the HEO architecture and the one that produces the longest-lasting results. It governs whether an entity's knowledge and authority are embedded in the parametric memory of large language models — the trained understanding that AI systems draw on when composing responses without consulting a retrieval layer. Parametric embedding cannot be engineered directly. It is the result of an entity's content being present in the training corpora of major AI models, which means the content must be published, indexed, authoritative, and semantically rich enough to be included in the datasets that models are trained on. The GEO layer requires five specific signal types that the SEO and AEO layers alone do not produce: documented specific outcomes, comparative differentiation, social proof architecture, authority positioning evidence, and entity completeness. Each of these signal types addresses a different aspect of the question that generative AI systems are implicitly asking when they decide whether to include an entity in a synthesized response: Is this entity real? Is it authoritative? Is it better than the alternatives? Can I cite it confidently? The GEO layer answers all five questions with documented, verifiable evidence.

Measurement is not the final phase of HEO in the sense that it comes after the work is done. It is the ongoing operational discipline that makes the work accountable, directs the next iteration, and provides the evidence that the HEO architecture is functioning as designed. The six core HEO metrics are not vanity metrics. Each one measures a specific, meaningful aspect of the entity's presence across the three-layer discovery architecture. Entity Representation Score measures the quality of how AI systems describe the entity — from absent (0) to cited as the primary recommendation (5). Platform Coverage Rate measures the breadth of the entity's presence — how many AI platforms mention it out of the total platforms tested. Citation Frequency measures how often the entity appears in AI-generated responses across a defined set of queries. Citation Accuracy Rate measures whether the AI's description of the entity is factually correct. Recommendation Rate measures how often the entity is actively recommended rather than merely mentioned. Citation Favorability Score measures whether the entity is positioned positively, neutrally, or negatively in AI-generated responses. Together, these six metrics provide a complete picture of the entity's AI Visibility status — and a clear signal of where the next iteration of HEO work should be directed.