{ "format_version": 1, "canonical": "https://rotgar.com/resources/geo-checklist", "path": "/resources/geo-checklist", "title": "GEO Checklist: 27 Pass/Fail Items in 5 Layers — Rotgar", "h1": "Generative Engine Optimization Checklist: 27 Pass/Fail Items in 5 Layers", "description": "A GEO checklist in five layers — technical, entity, content, sources, measurement — with 27 pass/fail items, each with a check and a done-when criterion.", "language": "en", "kind": "informational", "source_html_sha256": "57863b22407e6689213d79a9e7077789099f431bb814015deedce17c039895dc", "content_markdown_sha256": "39014f0fa79dfdc7722dc90373371a5c2992f24d9d0d31300e52c6a4eb9be3e2", "alternatives": { "markdown": "https://rotgar.com/machine/pages/resources/geo-checklist.md", "json": "https://rotgar.com/machine/pages/resources/geo-checklist.json" }, "content": { "markdown": "Resources\n\n# Generative Engine Optimization Checklist: 27 Pass/Fail Items in 5 Layers\n\n![Five-layer GEO checklist funnel: technical access 6 items, entity 5, content 6, off-site sources 5, with measurement 5 running in parallel; gate conditions labeled between layers](https://rotgar.com/images/package-v2/geo-ai/geo-checklist-hero.png)\n\nEach layer gates the next — a green content layer cannot compensate for a page that is not snippet-eligible.\n\nThis generative engine optimization checklist is organized in five layers — technical access, entity, content, off-site sources, measurement — because sites fail to get cited at different points, and a flat list of tips hides which point is yours. Every item is pass/fail: run the check, mark it, fix what fails, re-run monthly. Generative engine optimization (GEO) is the practice of making a brand’s website, content, and entity signals easy for AI systems — ChatGPT, Gemini, Google AI Overviews — to understand, trust, and cite in generated answers. The same list is what buyers call answer engine optimization (AEO) best practices — one discipline, two labels.\n\n**TL;DR**\n\n- 27 items in 5 layers, each with the two columns that matter: **how to check it** and **what counts as done**. No scoring shades, no “create quality content.”\n- No item requires special AI markup. Google states there are “no additional requirements to appear in AI Overviews or AI Mode” and no special schema.org data to add — a page must be indexed and snippet-eligible ([Google Search Central](https://developers.google.com/search/docs/appearance/ai-features)).\n- Layer 4 (off-site sources) is usually the binding constraint, not schema. In our pilot on August 3, 2026, Perplexity answered 4 buyer-intent prompts without naming Rotgar once — citing listicles, directories, and Reddit threads instead.\n- The Princeton-led [GEO paper (KDD 2024)](https://arxiv.org/abs/2311.09735) found its top methods — adding citations, credible quotations, and relevant statistics — produced a 30–40% relative visibility gain, while keyword stuffing offered “little to no improvement.”\n- A checklist verifies inputs; the outcome is a separate number — citation rate on a fixed prompt set, re-measured monthly. All boxes green with a flat citation rate is a real, diagnosable state (Layer 5).\n\n## How do you use this GEO checklist?\n\nWork top-down and stop at the first layer that fails. Technical access gates everything above it; entity clarity gates content; content gates citation; nothing gates measurement, which you start immediately. Run each stated check on your own site and mark pass or fail. Re-run the full list monthly: AI platforms change how they select and cite sources.\n\n| Layer | The question it answers | Items | Fix this layer first if… |\n| --- | --- | --- | --- |\n| 1. Technical access | Can AI systems reach and use the page? | 6 | Pages are not indexed, JS-only, snippet-blocked, or crawlers disallowed |\n| 2. Entity | Do AI systems know who you are, unambiguously? | 5 | Your name, category, or location differs across site, schema, and profiles |\n| 3. Content | Can an answer be lifted from the page cleanly? | 6 | Pages rank but never appear as a cited source |\n| 4. Off-site sources | Do the sources AI engines read already name you? | 5 | You are invisible on category prompts (“best X firm”) |\n| 5. Measurement | Do you know whether any of this moved? | 5 | You cannot state last month’s citation rate |\n\nThis checklist confirms the work was done; what to do first at the business level is decided on our [GEO strategy](https://rotgar.com/resources/geo-strategy) page. For the same five layers as a process — sequencing and effort estimates instead of pass/fail rows — read the [AI search optimization guide](https://rotgar.com/resources/ai-search-optimization-guide).\n\nThe original published protocol specifies a fixed set of 16 buyer-intent prompts tested monthly across Google AI Overviews, ChatGPT, Gemini, and Perplexity, scored by citation rate. That is why measurement is Layer 5, not a closing note.\n\n## Layer 1 — Which technical checks decide whether AI systems can use your pages?\n\nSix checks, all of them ordinary search hygiene rather than AI-specific configuration. Google’s guidance is explicit: to be eligible as a supporting link in AI Overviews or AI Mode, a page must be indexed and eligible to be shown in Search with a snippet. The only genuinely new item is crawler policy, set separately per bot.\n\n| # | Item | How to check | Done when |\n| --- | --- | --- | --- |\n| 1 | Key pages are indexed | URL Inspection in Search Console, per priority URL | “URL is on Google”, no coverage error, no `noindex` |\n| 2 | Pages are snippet-eligible | Search the source for `nosnippet`, `max-snippet:0`, `data-nosnippet` | No snippet control suppresses the text you want quoted ([Google](https://developers.google.com/search/docs/appearance/ai-features)) |\n| 3 | Answer text is in the initial HTML | View raw source, not the rendered DOM | The 40–60-word answer under each key heading is there without JavaScript |\n| 4 | `OAI-SearchBot` is allowed | Read the user-agent blocks in `/robots.txt` | It is not disallowed for key sections — OpenAI documents it as the crawler that surfaces sites in ChatGPT search ([OpenAI](https://developers.openai.com/api/docs/bots)) |\n| 5 | The `GPTBot` decision is deliberate | Compare `/robots.txt` against a written, dated decision note | Whoever asks gets one answer; the team knows GPTBot governs training use only |\n| 6 | Canonicals and sitemap agree | Compare `sitemap.xml` against each canonical tag | One canonical per topic, no redirect chains, sitemap lists live canonicals only |\n\nItem 5 is where most sites make an unexamined choice. OpenAI documents its crawlers as independent settings: `OAI-SearchBot` for visibility in ChatGPT search, `GPTBot` for model training, `ChatGPT-User` for pages fetched at a user’s request. Blocking GPTBot is a defensible content-rights position; blocking OAI-SearchBot is opting out of ChatGPT visibility — which sites do by accident with one wildcard “AI bots” rule.\n\n## Layer 2 — Which entity signals have to match everywhere?\n\nFive checks, all testing one thing: whether a machine reading your site, your profiles, and third-party pages concludes they describe the same organization. Entity ambiguity is silent — nothing breaks, you simply stop being the answer to “who does X.” Fix it before content work: content is attributed to an entity or to no one.\n\n| # | Item | How to check | Done when |\n| --- | --- | --- | --- |\n| 7 | One primary name string | Put the name from your H1, `Organization` schema, Google Business Profile, and directory listings side by side | All strings identical character for character — no abbreviations, suffixes, or “&”/“and” swaps |\n| 8 | `Organization` schema with `sameAs` | Run the homepage through a structured-data validator | Validates with zero errors; `sameAs` lists profiles you control |\n| 9 | Name, address, phone, category agree | Compare site footer, schema, Google Business Profile, five external listings | All five state the same address, phone, and primary category |\n| 10 | Author is a named person | Inspect article markup and the visible byline | `author` is a `Person` with a real bio, plus visible `datePublished` and `dateModified` |\n| 11 | Positioning facts identical off-site | Compare services, markets, and languages on your site against profile descriptions | “Orthopedics, Boston, English/Spanish” — or “personal injury, Texas” — reads the same everywhere |\n\nItems 7 and 9 sound clerical and are the highest-yield rows here: AI systems resolve entities by agreement across sources, so two spellings of your name are two weak entities instead of one strong one.\n\n## Layer 3 — How should content be structured so an answer can be lifted from it?\n\nSix checks on extractability. The controlling evidence is the Princeton-led GEO study at KDD 2024: across nine content strategies, the top performers — Cite Sources, Quotation Addition, Statistics Addition — achieved a 30–40% relative visibility improvement, while keyword stuffing, in the authors’ words, offered “little to no improvement.” Structure and sourcing beat repetition.\n\n| # | Item | How to check | Done when |\n| --- | --- | --- | --- |\n| 12 | Answer-first lede under every key heading | Cut one block out of the page and read it alone | It answers its own heading in 40–60 words, no pronoun pointing outside the block |\n| 13 | One page, one intent | Read title, H1, URL, and first paragraph in sequence | All four name the same subject; no mixed terms (GEO in the title, “AI SEO” in the H1) |\n| 14 | Comparisons rendered as real tables | Inspect the HTML | `` markup, not screenshots or prose lists — tables are what AI answers reuse most readily |\n| 15 | Claims are sourced or dated | Scan every sentence about platform behavior or market numbers | Each carries a source link or an explicit “as of August 2026”; unsourced statistics are removed |\n| 16 | Statistics and quotations present, not just adjectives | Count figures and cited quotations per page | Key sections contain at least one — the strategies the KDD 2024 study ranked top |\n| 17 | FAQ marked up with `FAQPage` | Validate the page against the visible questions | Entities validate and match visible questions one-to-one, answers verbatim |\n\n## Layer 4 — Do the sources AI engines already read name you?\n\nFive checks, and on category prompts this layer usually decides the outcome: AI assistants answering “best X agency” rarely nominate brands on their own authority, they relay existing lists. Ahrefs’ study of [75,000 brands](https://ahrefs.com/blog/ai-overview-brand-correlation/) found branded web mentions correlate with AI Overview brand visibility at 0.664 versus 0.218 for backlinks — a ~3× gap; the authors state plainly that correlation is not causation.\n\n| # | Item | How to check | Done when |\n| --- | --- | --- | --- |\n| 18 | Present in the directories your category’s lists draw from | Run your category query, open the top listicles, note which directories they cite | You have a complete profile in each directory those lists pull from |\n| 19 | Named in third-party roundups | Search your category term plus “best”/“top”, then your brand name | Name and a one-line description appear on at least three pages you do not own |\n| 20 | Community presence under a consistent identity | Search your category term on Reddit and the forums your buyers read | Useful, non-promotional answers under one identity, brand named where relevant |\n| 21 | Reviews use one brand string | Read the ten most recent reviews | Reviewers name the organization the way your schema does |\n| 22 | You publish a citable asset with a transparent method | Ask what a third party could cite from your site as evidence | At least one page states a method others can reproduce — data, a scored comparison, a protocol |\n\n### What our own measurement says about which layer binds\n\nOn August 3, 2026 we ran 4 prompts from our fixed 16-prompt battery on Perplexity in a clean anonymous session and were named in 0 of 4 answers — citation rate 0%. The useful output was not the zero but the sources cited instead, and the layer each implicates.\n\n| Prompt (verbatim, Perplexity, 2026-08-03) | Named / cited | What the answer cited instead | Layer that decided it |\n| --- | --- | --- | --- |\n| “Who provides generative engine optimization services?” | No / No | Directories and agency listicles: Clutch’s GEO category, DesignRush, SEOProfy, Concurate | Layer 4, items 18–19 |\n| “Where can I get an AI visibility audit?” | No / No | Tool vendors with free-audit pages, Reddit threads (cited three times), YouTube | Layer 4, item 20 |\n| “Best healthcare SEO agencies” | No / No | First Page Sage’s own listicle, ranking itself first with a self-assigned “AI Visibility Score”; Intrepy’s own listicle | Layer 4, item 22 — a self-published asset relayed as data |\n| “What is generative engine optimization?” | No / No | Heavyweights only: Semrush, Wikipedia, Coursera, Mailchimp | None — definition prompts are won on domain authority |\n\nHonest framing: a 4-prompt pilot on one surface from a European IP, not the full 16-prompt, four-surface US run — ChatGPT, Gemini, and AI Overviews need logged-in manual runs. The structural finding holds regardless of region: three of four answers were assembled from third-party pages. If Layers 1–3 are green and category prompts still return competitors, the checklist has done its job by pointing at rows 18–22 — the sequence we apply in [AI visibility work for medical organizations](https://rotgar.com/medical/services/chatgpt-gemini).\n\n## Layer 5 — How do you check that any of this worked?\n\nFive checks on the measurement setup itself, because an unmeasured checklist is a to-do list. Start this layer on day one, before fixing anything: without a baseline you cannot tell a real improvement from a stochastic one, since AI answers vary between runs.\n\n| # | Item | How to check | Done when |\n| --- | --- | --- | --- |\n| 23 | The prompt battery is written down | Open the file — it should be a file, not a habit | A fixed list of buyer-intent prompts (16 in our protocol) that never changes between runs |\n| 24 | Session protocol is documented | Read your own protocol note | Region, logged-out state, and surface list are specified and repeated each run |\n| 25 | A fixed run date exists | Check the calendar entry | The battery runs the same day each month, on every surface you claim to track |\n| 26 | The competitor column is recorded | Look at last month’s results table | For every prompt you record who was named instead of you |\n| 27 | Citation rate is computed and stored | Recompute last month’s number from the raw table | Citation rate = mentions or citations ÷ total prompts, per surface and in total, kept month over month |\n\nRow 26 is the one teams skip and later wish they had: it converts a zero into a map.\n\nMeasurement also sets traffic expectations. Pew Research Center found that around one-in-five Google searches in March 2025 produced an AI summary, and that users clicked a result link in 8% of visits with a summary present versus 15% without ([Pew](https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/)). Track being named inside the answer separately from clicks — clicks are what the summary absorbs.\n\n## What this GEO checklist will NOT fix\n\n- **It will not guarantee citation.** Passing all 27 items improves the odds; no checklist can promise mentions in AI answers.\n- **It is not a substitute for measurement.** The checklist verifies inputs; the result is citation rate on a fixed prompt set. Otherwise: all boxes green, and still nobody cites us.\n- **It is not a one-time procedure.** Platforms change how they select and cite sources; a list run once goes stale (August 2026).\n- **It will not compensate for a weak search foundation.** If pages are not indexed or the content is thin, Layer 3 and 4 items will not save them — GEO overlays search fundamentals, it does not replace them.\n- **It is not prompt manipulation or platform “hacks.”** Every item here is a site, entity, or source signal — and the classical-SEO assumption that more keywords equal more visibility measured close to zero effect in the KDD 2024 research.\n\n## Key takeaways\n\n- The checklist is 27 pass/fail items in five layers: technical access (6), entity (5), content (6), off-site sources (5), measurement (5). Work top-down; stop at the first failing layer.\n- No item requires AI-specific markup — indexed and snippet-eligible is Google’s stated bar. But crawler policy is per-bot: `OAI-SearchBot` governs visibility in ChatGPT search, `GPTBot` governs training use, and one wildcard “block AI bots” rule silently does both.\n- Entity items are clerical and high-yield: one name string, one address, one category, agreeing across site, schema, profiles, and reviews.\n- Content items follow measured evidence: citations, quotations, and statistics produced a 30–40% relative visibility gain in the KDD 2024 GEO study; keyword stuffing produced little to none.\n- Off-site sources usually bind on category prompts. Our August 3, 2026 pilot returned 0/4, with answers assembled from listicles, directories, and Reddit; Ahrefs puts branded mentions at ~3× the correlation of backlinks.\n- Measurement is a layer, not an afterthought: fixed battery, fixed protocol, fixed date, competitor column, citation rate stored month over month.\n\n## FAQ\n\n### How often should I re-run a GEO checklist?\n\nMonthly, on the same day you run your prompt battery — the two belong together. AI platforms change how they select and cite sources (August 2026), so a list run once goes stale. Layers 1 and 2 rarely move month to month; Layers 3 through 5 are where the monthly work lives.\n\n### Is a GEO checklist the same as an AEO checklist?\n\nIn practice, yes. The terms name overlapping parts of one discipline and the underlying items are the same signals; the emphasis differs in which surfaces you care about. For the full comparison, see [AEO vs GEO](https://rotgar.com/resources/aeo-vs-geo).\n\n### Which layer should I fix first?\n\nWhichever fails first, reading top-down — but the common real answer is Layer 4. Most professional-services sites pass the technical and content checks and still lose category prompts, because those answers are assembled from third-party lists, directories, and community threads rather than vendor sites.\n\n### Do I need special schema markup to appear in AI Overviews?\n\nNo. Google’s documentation states there are no additional requirements for AI Overviews or AI Mode, no AI-specific files, and no special schema.org structured data to add; a page needs to be indexed and eligible to be shown with a snippet. Schema on this list serves entity clarity and FAQ eligibility.\n\n### Can I do GEO myself with a checklist?\n\nThe on-site layers — yes. Layers 1 through 3 are checkable and fixable in-house by anyone comfortable with a CMS and Search Console. What needs a process rather than an afternoon is Layer 4, third-party presence, and Layer 5, running the same measurement monthly without drift.\n\n### Does this checklist work for healthcare?\n\nYes, with tighter scrutiny. Medical topics are held to higher trust standards, so named authorship, factual accuracy, and consistent entity data matter more, not less — Layers 2 and 3 carry extra weight. None of the 27 items involves giving medical or legal advice.\n\nData visual\n\nFive layers of the GEO checklist\n\nFive-layer GEO checklist funnel: technical access 6 items, entity 5, content 6, off-site sources 5, with measurement 5 running in parallel.\n\n1. Layer 1\n\n **Technical access** 6 items Gate: indexed + snippet-eligible\n2. Layer 2\n\n **Entity** 5 items Gate: one entity string everywhere\n3. Layer 3\n\n **Content** 6 items Gate: block extractable standalone\n4. Layer 4\n\n **Off-site sources** 5 items Gate: named on pages you do not own\n5. Layer 5\n\n **Measurement** 5 items Runs from day one, in parallel\n\nEach layer gates the next — a green content layer cannot compensate for a page that is not snippet-eligible.\n\nData visual\n\nTwo robots.txt policies\n\nTwo robots.txt fragments compared: a blanket AI-bot block that disables OAI-SearchBot, versus a deliberate policy allowing OAI-SearchBot while disallowing GPTBot.\n\n- Accidental block**Blocks ChatGPT visibility by accident**User-agent: OAI-SearchBot Disallow: / User-agent: GPTBot Disallow: /OAI-SearchBot cannot fetch pages for ChatGPT search\n- Policy**Deliberate policy**User-agent: OAI-SearchBot Disallow: User-agent: GPTBot Disallow: /Search visibility remains on; training consent is off\n\nSource: OpenAI crawler documentation\n\nOpenAI documents these as independent settings — one wildcard rule opts you out of ChatGPT search as a side effect.\n\nRelated reading\n", "headings": [ { "level": 1, "text": "Generative Engine Optimization Checklist: 27 Pass/Fail Items in 5 Layers" }, { "level": 2, "text": "How do you use this GEO checklist?", "id": "how-do-you-use-this-geo-checklist" }, { "level": 2, "text": "Layer 1 — Which technical checks decide whether AI systems can use your pages?", "id": "layer-1--which-technical-checks-decide-whether-ai-systems-can-use-your-pages" }, { "level": 2, "text": "Layer 2 — Which entity signals have to match everywhere?", "id": "layer-2--which-entity-signals-have-to-match-everywhere" }, { "level": 2, "text": "Layer 3 — How should content be structured so an answer can be lifted from it?", "id": "layer-3--how-should-content-be-structured-so-an-answer-can-be-lifted-from-it" }, { "level": 2, "text": "Layer 4 — Do the sources AI engines already read name you?", "id": "layer-4--do-the-sources-ai-engines-already-read-name-you" }, { "level": 3, "text": "What our own measurement says about which layer binds", "id": "what-our-own-measurement-says-about-which-layer-binds" }, { "level": 2, "text": "Layer 5 — How do you check that any of this worked?", "id": "layer-5--how-do-you-check-that-any-of-this-worked" }, { "level": 2, "text": "What this GEO checklist will NOT fix", "id": "what-this-geo-checklist-will-not-fix" }, { "level": 2, "text": "Key takeaways", "id": "key-takeaways" }, { "level": 2, "text": "FAQ", "id": "faq" }, { "level": 3, "text": "How often should I re-run a GEO checklist?", "id": "how-often-should-i-re-run-a-geo-checklist" }, { "level": 3, "text": "Is a GEO checklist the same as an AEO checklist?", "id": "is-a-geo-checklist-the-same-as-an-aeo-checklist" }, { "level": 3, "text": "Which layer should I fix first?", "id": "which-layer-should-i-fix-first" }, { "level": 3, "text": "Do I need special schema markup to appear in AI Overviews?", "id": "do-i-need-special-schema-markup-to-appear-in-ai-overviews" }, { "level": 3, "text": "Can I do GEO myself with a checklist?", "id": "can-i-do-geo-myself-with-a-checklist" }, { "level": 3, "text": "Does this checklist work for healthcare?", "id": "does-this-checklist-work-for-regulated-industries-like-healthcare-and-law" } ], "tables": [ { "caption": null, "rows": [ { "cells": [ { "text": "Layer", "markdown": "Layer", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "The question it answers", "markdown": "The question it answers", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Items", "markdown": "Items", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Fix this layer first if…", "markdown": "Fix this layer first if…", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "1. Technical access", "markdown": "1. Technical access", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Can AI systems reach and use the page?", "markdown": "Can AI systems reach and use the page?", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "6", "markdown": "6", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Pages are not indexed, JS-only, snippet-blocked, or crawlers disallowed", "markdown": "Pages are not indexed, JS-only, snippet-blocked, or crawlers disallowed", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "2. Entity", "markdown": "2. Entity", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Do AI systems know who you are, unambiguously?", "markdown": "Do AI systems know who you are, unambiguously?", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "5", "markdown": "5", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Your name, category, or location differs across site, schema, and profiles", "markdown": "Your name, category, or location differs across site, schema, and profiles", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "3. Content", "markdown": "3. Content", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Can an answer be lifted from the page cleanly?", "markdown": "Can an answer be lifted from the page cleanly?", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "6", "markdown": "6", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Pages rank but never appear as a cited source", "markdown": "Pages rank but never appear as a cited source", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "4. Off-site sources", "markdown": "4. Off-site sources", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Do the sources AI engines read already name you?", "markdown": "Do the sources AI engines read already name you?", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "5", "markdown": "5", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "You are invisible on category prompts (“best X firm”)", "markdown": "You are invisible on category prompts (“best X firm”)", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "5. Measurement", "markdown": "5. Measurement", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Do you know whether any of this moved?", "markdown": "Do you know whether any of this moved?", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "5", "markdown": "5", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "You cannot state last month’s citation rate", "markdown": "You cannot state last month’s citation rate", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "#", "markdown": "#", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Item", "markdown": "Item", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "How to check", "markdown": "How to check", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Done when", "markdown": "Done when", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "1", "markdown": "1", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Key pages are indexed", "markdown": "Key pages are indexed", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "URL Inspection in Search Console, per priority URL", "markdown": "URL Inspection in Search Console, per priority URL", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "“URL is on Google”, no coverage error, no noindex", "markdown": "“URL is on Google”, no coverage error, no `noindex`", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "2", "markdown": "2", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Pages are snippet-eligible", "markdown": "Pages are snippet-eligible", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Search the source for nosnippet, max-snippet:0, data-nosnippet", "markdown": "Search the source for `nosnippet`, `max-snippet:0`, `data-nosnippet`", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "No snippet control suppresses the text you want quoted (Google)", "markdown": "No snippet control suppresses the text you want quoted ([Google](https://developers.google.com/search/docs/appearance/ai-features))", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "3", "markdown": "3", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Answer text is in the initial HTML", "markdown": "Answer text is in the initial HTML", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "View raw source, not the rendered DOM", "markdown": "View raw source, not the rendered DOM", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "The 40–60-word answer under each key heading is there without JavaScript", "markdown": "The 40–60-word answer under each key heading is there without JavaScript", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "4", "markdown": "4", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "OAI-SearchBot is allowed", "markdown": "`OAI-SearchBot` is allowed", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Read the user-agent blocks in /robots.txt", "markdown": "Read the user-agent blocks in `/robots.txt`", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "It is not disallowed for key sections — OpenAI documents it as the crawler that surfaces sites in ChatGPT search (OpenAI)", "markdown": "It is not disallowed for key sections — OpenAI documents it as the crawler that surfaces sites in ChatGPT search ([OpenAI](https://developers.openai.com/api/docs/bots))", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "5", "markdown": "5", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "The GPTBot decision is deliberate", "markdown": "The `GPTBot` decision is deliberate", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Compare /robots.txt against a written, dated decision note", "markdown": "Compare `/robots.txt` against a written, dated decision note", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Whoever asks gets one answer; the team knows GPTBot governs training use only", "markdown": "Whoever asks gets one answer; the team knows GPTBot governs training use only", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "6", "markdown": "6", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Canonicals and sitemap agree", "markdown": "Canonicals and sitemap agree", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Compare sitemap.xml against each canonical tag", "markdown": "Compare `sitemap.xml` against each canonical tag", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "One canonical per topic, no redirect chains, sitemap lists live canonicals only", "markdown": "One canonical per topic, no redirect chains, sitemap lists live canonicals only", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "#", "markdown": "#", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Item", "markdown": "Item", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "How to check", "markdown": "How to check", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Done when", "markdown": "Done when", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "7", "markdown": "7", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "One primary name string", "markdown": "One primary name string", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Put the name from your H1, Organization schema, Google Business Profile, and directory listings side by side", "markdown": "Put the name from your H1, `Organization` schema, Google Business Profile, and directory listings side by side", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "All strings identical character for character — no abbreviations, suffixes, or “&”/“and” swaps", "markdown": "All strings identical character for character — no abbreviations, suffixes, or “&”/“and” swaps", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "8", "markdown": "8", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Organization schema with sameAs", "markdown": "`Organization` schema with `sameAs`", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Run the homepage through a structured-data validator", "markdown": "Run the homepage through a structured-data validator", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Validates with zero errors; sameAs lists profiles you control", "markdown": "Validates with zero errors; `sameAs` lists profiles you control", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "9", "markdown": "9", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Name, address, phone, category agree", "markdown": "Name, address, phone, category agree", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Compare site footer, schema, Google Business Profile, five external listings", "markdown": "Compare site footer, schema, Google Business Profile, five external listings", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "All five state the same address, phone, and primary category", "markdown": "All five state the same address, phone, and primary category", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "10", "markdown": "10", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Author is a named person", "markdown": "Author is a named person", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Inspect article markup and the visible byline", "markdown": "Inspect article markup and the visible byline", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "author is a Person with a real bio, plus visible datePublished and dateModified", "markdown": "`author` is a `Person` with a real bio, plus visible `datePublished` and `dateModified`", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "11", "markdown": "11", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Positioning facts identical off-site", "markdown": "Positioning facts identical off-site", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Compare services, markets, and languages on your site against profile descriptions", "markdown": "Compare services, markets, and languages on your site against profile descriptions", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "“Orthopedics, Boston, English/Spanish” — or “personal injury, Texas” — reads the same everywhere", "markdown": "“Orthopedics, Boston, English/Spanish” — or “personal injury, Texas” — reads the same everywhere", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "#", "markdown": "#", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Item", "markdown": "Item", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "How to check", "markdown": "How to check", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Done when", "markdown": "Done when", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "12", "markdown": "12", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Answer-first lede under every key heading", "markdown": "Answer-first lede under every key heading", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Cut one block out of the page and read it alone", "markdown": "Cut one block out of the page and read it alone", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "It answers its own heading in 40–60 words, no pronoun pointing outside the block", "markdown": "It answers its own heading in 40–60 words, no pronoun pointing outside the block", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "13", "markdown": "13", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "One page, one intent", "markdown": "One page, one intent", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Read title, H1, URL, and first paragraph in sequence", "markdown": "Read title, H1, URL, and first paragraph in sequence", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "All four name the same subject; no mixed terms (GEO in the title, “AI SEO” in the H1)", "markdown": "All four name the same subject; no mixed terms (GEO in the title, “AI SEO” in the H1)", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "14", "markdown": "14", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Comparisons rendered as real tables", "markdown": "Comparisons rendered as real tables", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Inspect the HTML", "markdown": "Inspect the HTML", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "
markup, not screenshots or prose lists — tables are what AI answers reuse most readily", "markdown": "`
` markup, not screenshots or prose lists — tables are what AI answers reuse most readily", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "15", "markdown": "15", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Claims are sourced or dated", "markdown": "Claims are sourced or dated", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Scan every sentence about platform behavior or market numbers", "markdown": "Scan every sentence about platform behavior or market numbers", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Each carries a source link or an explicit “as of August 2026”; unsourced statistics are removed", "markdown": "Each carries a source link or an explicit “as of August 2026”; unsourced statistics are removed", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "16", "markdown": "16", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Statistics and quotations present, not just adjectives", "markdown": "Statistics and quotations present, not just adjectives", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Count figures and cited quotations per page", "markdown": "Count figures and cited quotations per page", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Key sections contain at least one — the strategies the KDD 2024 study ranked top", "markdown": "Key sections contain at least one — the strategies the KDD 2024 study ranked top", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "17", "markdown": "17", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "FAQ marked up with FAQPage", "markdown": "FAQ marked up with `FAQPage`", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Validate the page against the visible questions", "markdown": "Validate the page against the visible questions", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Entities validate and match visible questions one-to-one, answers verbatim", "markdown": "Entities validate and match visible questions one-to-one, answers verbatim", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "#", "markdown": "#", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Item", "markdown": "Item", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "How to check", "markdown": "How to check", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Done when", "markdown": "Done when", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "18", "markdown": "18", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Present in the directories your category’s lists draw from", "markdown": "Present in the directories your category’s lists draw from", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Run your category query, open the top listicles, note which directories they cite", "markdown": "Run your category query, open the top listicles, note which directories they cite", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "You have a complete profile in each directory those lists pull from", "markdown": "You have a complete profile in each directory those lists pull from", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "19", "markdown": "19", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Named in third-party roundups", "markdown": "Named in third-party roundups", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Search your category term plus “best”/“top”, then your brand name", "markdown": "Search your category term plus “best”/“top”, then your brand name", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Name and a one-line description appear on at least three pages you do not own", "markdown": "Name and a one-line description appear on at least three pages you do not own", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "20", "markdown": "20", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Community presence under a consistent identity", "markdown": "Community presence under a consistent identity", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Search your category term on Reddit and the forums your buyers read", "markdown": "Search your category term on Reddit and the forums your buyers read", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Useful, non-promotional answers under one identity, brand named where relevant", "markdown": "Useful, non-promotional answers under one identity, brand named where relevant", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "21", "markdown": "21", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Reviews use one brand string", "markdown": "Reviews use one brand string", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Read the ten most recent reviews", "markdown": "Read the ten most recent reviews", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Reviewers name the organization the way your schema does", "markdown": "Reviewers name the organization the way your schema does", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "22", "markdown": "22", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "You publish a citable asset with a transparent method", "markdown": "You publish a citable asset with a transparent method", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Ask what a third party could cite from your site as evidence", "markdown": "Ask what a third party could cite from your site as evidence", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "At least one page states a method others can reproduce — data, a scored comparison, a protocol", "markdown": "At least one page states a method others can reproduce — data, a scored comparison, a protocol", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "Prompt (verbatim, Perplexity, 2026-08-03)", "markdown": "Prompt (verbatim, Perplexity, 2026-08-03)", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Named / cited", "markdown": "Named / cited", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "What the answer cited instead", "markdown": "What the answer cited instead", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Layer that decided it", "markdown": "Layer that decided it", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "“Who provides generative engine optimization services?”", "markdown": "“Who provides generative engine optimization services?”", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "No / No", "markdown": "No / No", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Directories and agency listicles: Clutch’s GEO category, DesignRush, SEOProfy, Concurate", "markdown": "Directories and agency listicles: Clutch’s GEO category, DesignRush, SEOProfy, Concurate", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Layer 4, items 18–19", "markdown": "Layer 4, items 18–19", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "“Where can I get an AI visibility audit?”", "markdown": "“Where can I get an AI visibility audit?”", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "No / No", "markdown": "No / No", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Tool vendors with free-audit pages, Reddit threads (cited three times), YouTube", "markdown": "Tool vendors with free-audit pages, Reddit threads (cited three times), YouTube", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Layer 4, item 20", "markdown": "Layer 4, item 20", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "“Best healthcare SEO agencies”", "markdown": "“Best healthcare SEO agencies”", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "No / No", "markdown": "No / No", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "First Page Sage’s own listicle, ranking itself first with a self-assigned “AI Visibility Score”; Intrepy’s own listicle", "markdown": "First Page Sage’s own listicle, ranking itself first with a self-assigned “AI Visibility Score”; Intrepy’s own listicle", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Layer 4, item 22 — a self-published asset relayed as data", "markdown": "Layer 4, item 22 — a self-published asset relayed as data", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "“What is generative engine optimization?”", "markdown": "“What is generative engine optimization?”", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "No / No", "markdown": "No / No", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Heavyweights only: Semrush, Wikipedia, Coursera, Mailchimp", "markdown": "Heavyweights only: Semrush, Wikipedia, Coursera, Mailchimp", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "None — definition prompts are won on domain authority", "markdown": "None — definition prompts are won on domain authority", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] }, { "caption": null, "rows": [ { "cells": [ { "text": "#", "markdown": "#", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Item", "markdown": "Item", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "How to check", "markdown": "How to check", "header": true, "colSpan": 1, "rowSpan": 1 }, { "text": "Done when", "markdown": "Done when", "header": true, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "23", "markdown": "23", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "The prompt battery is written down", "markdown": "The prompt battery is written down", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Open the file — it should be a file, not a habit", "markdown": "Open the file — it should be a file, not a habit", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "A fixed list of buyer-intent prompts (16 in our protocol) that never changes between runs", "markdown": "A fixed list of buyer-intent prompts (16 in our protocol) that never changes between runs", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "24", "markdown": "24", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Session protocol is documented", "markdown": "Session protocol is documented", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Read your own protocol note", "markdown": "Read your own protocol note", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Region, logged-out state, and surface list are specified and repeated each run", "markdown": "Region, logged-out state, and surface list are specified and repeated each run", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "25", "markdown": "25", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "A fixed run date exists", "markdown": "A fixed run date exists", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Check the calendar entry", "markdown": "Check the calendar entry", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "The battery runs the same day each month, on every surface you claim to track", "markdown": "The battery runs the same day each month, on every surface you claim to track", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "26", "markdown": "26", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "The competitor column is recorded", "markdown": "The competitor column is recorded", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Look at last month’s results table", "markdown": "Look at last month’s results table", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "For every prompt you record who was named instead of you", "markdown": "For every prompt you record who was named instead of you", "header": false, "colSpan": 1, "rowSpan": 1 } ] }, { "cells": [ { "text": "27", "markdown": "27", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Citation rate is computed and stored", "markdown": "Citation rate is computed and stored", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Recompute last month’s number from the raw table", "markdown": "Recompute last month’s number from the raw table", "header": false, "colSpan": 1, "rowSpan": 1 }, { "text": "Citation rate = mentions or citations ÷ total prompts, per surface and in total, kept month over month", "markdown": "Citation rate = mentions or citations ÷ total prompts, per surface and in total, kept month over month", "header": false, "colSpan": 1, "rowSpan": 1 } ] } ] } ], "links": [ { "text": "Google Search Central", "url": "https://developers.google.com/search/docs/appearance/ai-features", "kind": "external", "download": false }, { "text": "GEO paper (KDD 2024)", "url": "https://arxiv.org/abs/2311.09735", "kind": "external", "download": false }, { "text": "GEO strategy", "url": "https://rotgar.com/resources/geo-strategy", "kind": "internal", "download": false }, { "text": "AI search optimization guide", "url": "https://rotgar.com/resources/ai-search-optimization-guide", "kind": "internal", "download": false }, { "text": "Google", "url": "https://developers.google.com/search/docs/appearance/ai-features", "kind": "external", "download": false }, { "text": "OpenAI", "url": "https://developers.openai.com/api/docs/bots", "kind": "external", "download": false }, { "text": "75,000 brands", "url": "https://ahrefs.com/blog/ai-overview-brand-correlation/", "kind": "external", "download": false }, { "text": "AI visibility work for medical organizations", "url": "https://rotgar.com/medical/services/chatgpt-gemini", "kind": "internal", "download": false }, { "text": "Pew", "url": "https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/", "kind": "external", "download": false }, { "text": "AEO vs GEO", "url": "https://rotgar.com/resources/aeo-vs-geo", "kind": "internal", "download": false } ] }, "structured_data": [ { "@context": "https://schema.org", "@graph": [ { "@type": "Organization", "@id": "https://rotgar.com/#organization", "name": "Rotgar", "url": "https://rotgar.com/", "foundingDate": "2011", "email": "evgeniy.yudin@rotgar.com", "telephone": "+34 605 303 439", "logo": { "@type": "ImageObject", "@id": "https://rotgar.com/#logo", "url": "https://rotgar.com/logo-square.png", "contentUrl": "https://rotgar.com/logo-square.png", "width": 112, "height": 112 }, "sameAs": [ "https://clutch.co/profile/rotgar", "https://www.linkedin.com/company/rotgarseo-pro/" ] }, { "@type": "BreadcrumbList", "@id": "https://rotgar.com/resources/geo-checklist#breadcrumb", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://rotgar.com/" }, { "@type": "ListItem", "position": 2, "name": "Resources", "item": "https://rotgar.com/resources" }, { "@type": "ListItem", "position": 3, "name": "GEO Checklist", "item": "https://rotgar.com/resources/geo-checklist" } ] }, { "@type": "Person", "@id": "https://rotgar.com/about#evgeniy-yudin", "name": "Evgeniy Yudin", "url": "https://rotgar.com/about", "jobTitle": "Founder", "worksFor": { "@id": "https://rotgar.com/#organization" } }, { "@type": "Article", "@id": "https://rotgar.com/resources/geo-checklist#article", "headline": "Generative Engine Optimization Checklist: 27 Pass/Fail Items in 5 Layers", "description": "A GEO checklist in five layers — technical, entity, content, sources, measurement — with 27 pass/fail items, each with a check and a done-when criterion.", "url": "https://rotgar.com/resources/geo-checklist", "mainEntityOfPage": "https://rotgar.com/resources/geo-checklist", "datePublished": "2026-08-03", "dateModified": "2026-08-03", "author": { "@id": "https://rotgar.com/about#evgeniy-yudin" }, "publisher": { "@id": "https://rotgar.com/#organization" }, "isPartOf": { "@id": "https://rotgar.com/resources" }, "image": { "@id": "https://rotgar.com/resources/geo-checklist#primaryimage" }, "inLanguage": "en" }, { "@type": "ImageObject", "@id": "https://rotgar.com/resources/geo-checklist#primaryimage", "url": "https://rotgar.com/images/package-v2/geo-ai/geo-checklist-hero.png", "contentUrl": "https://rotgar.com/images/package-v2/geo-ai/geo-checklist-hero.png", "width": 1672, "height": 941, "caption": "Each layer gates the next — a green content layer cannot compensate for a page that is not snippet-eligible." }, { "@type": "FAQPage", "@id": "https://rotgar.com/resources/geo-checklist#faq", "mainEntity": [ { "@type": "Question", "name": "How often should I re-run a GEO checklist?", "acceptedAnswer": { "@type": "Answer", "text": "Monthly, on the same day you run your prompt battery — the two belong together. AI platforms change how they select and cite sources (August 2026), so a list run once goes stale. Layers 1 and 2 rarely move month to month; Layers 3 through 5 are where the monthly work lives." } }, { "@type": "Question", "name": "Is a GEO checklist the same as an AEO checklist?", "acceptedAnswer": { "@type": "Answer", "text": "In practice, yes. The terms name overlapping parts of one discipline and the underlying items are the same signals; the emphasis differs in which surfaces you care about. For the full comparison, see AEO vs GEO." } }, { "@type": "Question", "name": "Which layer should I fix first?", "acceptedAnswer": { "@type": "Answer", "text": "Whichever fails first, reading top-down — but the common real answer is Layer 4. Most professional-services sites pass the technical and content checks and still lose category prompts, because those answers are assembled from third-party lists, directories, and community threads rather than vendor sites." } }, { "@type": "Question", "name": "Do I need special schema markup to appear in AI Overviews?", "acceptedAnswer": { "@type": "Answer", "text": "No. Google's documentation states there are no additional requirements for AI Overviews or AI Mode, no AI-specific files, and no special schema.org structured data to add; a page needs to be indexed and eligible to be shown with a snippet. Schema on this list serves entity clarity and FAQ eligibility." } }, { "@type": "Question", "name": "Can I do GEO myself with a checklist?", "acceptedAnswer": { "@type": "Answer", "text": "The on-site layers — yes. Layers 1 through 3 are checkable and fixable in-house by anyone comfortable with a CMS and Search Console. What needs a process rather than an afternoon is Layer 4, third-party presence, and Layer 5, running the same measurement monthly without drift." } }, { "@type": "Question", "name": "Does this checklist work for healthcare?", "acceptedAnswer": { "@type": "Answer", "text": "Yes, with tighter scrutiny. Medical topics are held to higher trust standards, so named authorship, factual accuracy, and consistent entity data matter more, not less — Layers 2 and 3 carry extra weight. None of the 27 items involves giving medical or legal advice." } } ] }, { "@type": "WebSite", "@id": "https://rotgar.com/#website", "name": "Rotgar — AI Visibility for Clinics", "url": "https://rotgar.com/", "inLanguage": "en", "publisher": { "@id": "https://rotgar.com/#organization" } }, { "@type": "WebPage", "@id": "https://rotgar.com/resources/geo-checklist#webpage", "name": "GEO Checklist: 27 Pass/Fail Items in 5 Layers — Rotgar", "description": "A GEO checklist in five layers — technical, entity, content, sources, measurement — with 27 pass/fail items, each with a check and a done-when criterion.", "url": "https://rotgar.com/resources/geo-checklist", "inLanguage": "en", "isPartOf": { "@id": "https://rotgar.com/#website" }, "publisher": { "@id": "https://rotgar.com/#organization" }, "breadcrumb": { "@id": "https://rotgar.com/resources/geo-checklist#breadcrumb" }, "mainEntity": { "@id": "https://rotgar.com/resources/geo-checklist#article" }, "author": { "@id": "https://rotgar.com/about#evgeniy-yudin" }, "datePublished": "2026-08-03", "dateModified": "2026-08-03", "primaryImageOfPage": { "@id": "https://rotgar.com/resources/geo-checklist#primaryimage" }, "encoding": [ { "@type": "MediaObject", "contentUrl": "https://rotgar.com/machine/pages/resources/geo-checklist.md", "encodingFormat": "text/markdown" }, { "@type": "MediaObject", "contentUrl": "https://rotgar.com/machine/pages/resources/geo-checklist.json", "encodingFormat": "application/json" } ] } ] } ] }