# CiteCheck — full guide text > All 11 CiteCheck guides on AI search visibility (GEO) in one file, for AI agents and LLMs. Each guide lists its sources. --- # AI crawlers list: GPTBot, OAI-SearchBot, PerplexityBot and more Source: https://citecheck.torontosoft.com/guides/ai-crawlers-robots-txt Updated: 2026-11-20 > AI companies use separate user agents for search, user-requested fetches and model training. For visibility in AI answers, allow the search crawlers (OAI-SearchBot, PerplexityBot, Claude-SearchBot, Googlebot, Bingbot, Applebot). Training crawlers (GPTBot, ClaudeBot, Google-Extended, Applebot-Extended) can be blocked without removing you from those search features. ## The AI crawler table Each company documents its own agents. Sorted by purpose: | User agent (robots.txt token) | Company | Purpose | Affects AI search visibility? | |---|---|---|---| | `OAI-SearchBot` | OpenAI | Surfaces sites in ChatGPT search answers | **Yes** | | `ChatGPT-User` | OpenAI | Fetches a page when a ChatGPT user asks; robots.txt may not apply | Indirectly | | `GPTBot` | OpenAI | Crawls content that may be used to train OpenAI models | No (training) | | `PerplexityBot` | Perplexity | Surfaces and links sites in Perplexity results; not used for training | **Yes** | | `Perplexity-User` | Perplexity | Fetches a page when a user asks; generally ignores robots.txt | Indirectly | | `Claude-SearchBot` | Anthropic | Indexes content to improve Claude's search results | **Yes** | | `Claude-User` | Anthropic | Fetches pages when a Claude user asks | Indirectly | | `ClaudeBot` | Anthropic | Collects content that may be used for training | No (training) | | `Googlebot` | Google | Google Search, including AI Overviews and AI Mode | **Yes** | | `Google-Extended` | Google | Gemini training and grounding in Gemini Apps / Vertex AI; not a Search signal | No for Search | | `Bingbot` | Microsoft | Bing Search | **Yes** | | `Applebot` | Apple | Siri, Spotlight and Safari search; may ground Apple AI answers | **Yes** | | `Applebot-Extended` | Apple | Controls use of Applebot data for training Apple models; does not crawl | No (training) | Sources: [OpenAI](https://platform.openai.com/docs/bots), [Perplexity](https://docs.perplexity.ai/guides/bots), [Anthropic](https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler), [Google](https://developers.google.com/search/docs/crawling-indexing/google-common-crawlers), [Bing](https://www.bing.com/webmasters/help/which-crawlers-does-bing-use-8c184ec0), [Apple](https://support.apple.com/en-us/119829). ## Three things people get wrong **1. Blocking training bots doesn't hide you from AI search.** OpenAI says its settings are independent: you can allow OAI-SearchBot to appear in search results while disallowing GPTBot. Google says Google-Extended "does not impact a site's inclusion in Google Search nor is it used as a ranking signal". Apple says pages that disallow Applebot-Extended can still be included in search results. **2. Blocking search bots does hide you.** OpenAI says sites opted out of OAI-SearchBot "will not be shown in ChatGPT search answers, though can still appear as navigational links". Perplexity recommends allowing PerplexityBot "to ensure your site appears in search results". Anthropic says disabling Claude-SearchBot "may reduce your site's visibility and accuracy in user search results". **3. User-triggered fetchers are different.** ChatGPT-User and Perplexity-User act when a person asks about a page. Both companies say robots.txt may not apply to them. If you need to block them, use your firewall. ## How robots.txt matching works Under the Robots Exclusion Protocol ([RFC 9309](https://www.rfc-editor.org/rfc/rfc9309)): - A crawler follows the group whose `User-agent` line **names it**. If no group names it, it follows the `User-agent: *` group. - Within a group, the **most specific (longest) matching rule** wins. If an `Allow` and a `Disallow` rule are equally specific, `Allow` wins. - An empty `Disallow:` line allows everything. Two vendor-specific details: Apple says that if robots.txt doesn't mention Applebot but does mention Googlebot, **Applebot follows the Googlebot rules**. Anthropic supports the non-standard `Crawl-delay` directive. A common accident: a site blocks everything for `*` to stop scrapers, and AI search bots that aren't named inherit that block. ## Copy-paste robots.txt rules **Allow AI search, opt out of training (a common middle ground):** ``` User-agent: OAI-SearchBot Allow: / User-agent: PerplexityBot Allow: / User-agent: Claude-SearchBot Allow: / User-agent: GPTBot Disallow: / User-agent: ClaudeBot Disallow: / User-agent: Google-Extended Disallow: / User-agent: Applebot-Extended Disallow: / User-agent: * Allow: / ``` **Allow everything (maximum visibility, including training):** ``` User-agent: * Allow: / Sitemap: https://www.example.com/sitemap.xml ``` Changes aren't instant: OpenAI cites about 24 hours for search, and Perplexity says up to 24 hours. ## Check your site in seconds The free [AI crawler checker](/tools/ai-crawler-checker) fetches your robots.txt and shows, for each agent above, whether it's allowed, blocked or partially blocked, and which group it matched. It applies the matching rules above, including Applebot's fallback to Googlebot rules. ## FAQ ### If I block GPTBot, will I disappear from ChatGPT? No. OpenAI says GPTBot and OAI-SearchBot are independent: disallowing GPTBot signals that content should not be used for training, while OAI-SearchBot controls whether your site appears in ChatGPT search answers. ### Does blocking Google-Extended remove me from AI Overviews? No. Google says Google-Extended does not affect inclusion in Google Search and is not a ranking signal. AI Overviews are controlled through Googlebot and snippet controls. ### Do user-triggered agents obey robots.txt? Not always. OpenAI says robots.txt rules may not apply to ChatGPT-User because actions are user-initiated, and Perplexity says Perplexity-User generally ignores robots.txt for the same reason. ### How fast do robots.txt changes take effect? OpenAI says it can take about 24 hours for search results to reflect a robots.txt update; Perplexity says up to 24 hours. ## Sources 1. [OpenAI — Overview of OpenAI crawlers](https://platform.openai.com/docs/bots) 2. [Perplexity — Perplexity crawlers](https://docs.perplexity.ai/guides/bots) 3. [Anthropic — Does Anthropic crawl data from the web, and how can site owners block the crawler?](https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler) 4. [Google — List of Google's common crawlers](https://developers.google.com/search/docs/crawling-indexing/google-common-crawlers) 5. [Apple Support — About Applebot](https://support.apple.com/en-us/119829) 6. [Bing Webmaster Tools — Which crawlers does Bing use?](https://www.bing.com/webmasters/help/which-crawlers-does-bing-use-8c184ec0) 7. [RFC 9309 — Robots Exclusion Protocol](https://www.rfc-editor.org/rfc/rfc9309) --- # Google AI Overviews vs SEO: what changes and what doesn't Source: https://citecheck.torontosoft.com/guides/ai-overviews-vs-seo Updated: 2026-11-20 > Google says there are no extra requirements or special optimizations for AI Overviews or AI Mode: a page must be indexed and eligible for a snippet, and normal SEO best practices apply. What changes is clicks. Pew found users clicked a result in 8% of visits when an AI summary appeared, versus 15% without one. ## The short version **Google's position:** "There are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary" ([Google Search Central](https://developers.google.com/search/docs/appearance/ai-features)). To be shown as a supporting link, a page must be **indexed and eligible to be shown with a snippet**. **What the data shows:** fewer clicks. In Pew Research Center's analysis of real browsing by 900 U.S. adults in March 2025, users clicked a traditional result in **8%** of visits to pages with an AI summary, versus **15%** without one. They clicked a link inside the summary itself in just **1%** of visits ([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/)). So SEO fundamentals still decide **whether** you can be included, while AI Overviews change **how much traffic** a given ranking produces. ## AI Overviews and SEO side by side | | Classic results | AI Overviews / AI Mode | |---|---|---| | Eligibility | Indexed page | Indexed **and** snippet-eligible page | | Extra markup required | No | No (per Google) | | How results are found | Query → ranked links | May use "query fan-out": multiple related searches across subtopics | | Reporting | Search Console Performance | Included in the same Performance report, "Web" search type | | Opt-out control | robots.txt, noindex | robots.txt for Googlebot, plus nosnippet, data-nosnippet, max-snippet, noindex | | How often shown | Every search | 18% of searches in Pew's March 2025 sample | Sources: [Google AI features documentation](https://developers.google.com/search/docs/appearance/ai-features); [Pew Research Center](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/). ## What "query fan-out" means for content Google says AI Overviews and AI Mode may issue **multiple related searches across subtopics** to build one answer, and can show a wider set of supporting links than a classic results page. For a site owner, that rewards pages that cover a **specific subtopic clearly**. A page that fully answers "how long does a furnace replacement take" can be pulled in as a supporting link for a broader question about furnace replacement. ## Which searches trigger AI summaries From Pew's data: - **18%** of all Google searches in the sample produced an AI summary. - Only **8%** of one- or two-word searches did, but **53%** of searches with 10 or more words did. - **60%** of searches starting with who, what, when or why produced one. - The median summary was **67 words**, and **88%** cited three or more sources. Long, question-style searches are where AI summaries appear most often. Those are also the searches people type into ChatGPT and Perplexity. ## What to change in an SEO plan 1. **Keep the fundamentals.** Google's own list: allow crawling in robots.txt *and* in your CDN or hosting, use internal links, provide a good page experience, keep important content in text, support it with images and video, and make structured data match the visible text. 2. **Don't block snippets by accident.** A page with `nosnippet` can't be a supporting link. Check templates for stray `max-snippet:0` or `data-nosnippet` on main content. 3. **Write for questions, not just keywords.** Answer the question in the first paragraph, then go deeper. 4. **Measure beyond clicks.** Search Console impressions still count when you're shown in an AI Overview. Use them, along with conversions, rather than raw sessions alone. 5. **Keep Merchant Center and Business Profile up to date.** Google lists this explicitly as a best practice for AI features. ## What Google-Extended does and doesn't do Blocking **Google-Extended** in robots.txt controls whether Google may use your content to train future Gemini models and for grounding in Gemini Apps and Vertex AI. Google says it **does not impact inclusion in Google Search and is not a ranking signal** ([Google's crawler list](https://developers.google.com/search/docs/crawling-indexing/google-common-crawlers)). To keep your content out of AI Overviews you would have to use the Googlebot and snippet controls above, which also affect regular Search. ## Beyond Google AI Overviews are one AI surface. ChatGPT, Perplexity and Gemini each pick sources differently. See [How to check if ChatGPT recommends your business](/guides/how-to-check-if-chatgpt-recommends-your-business), or run the [free AI visibility check](/check). ## FAQ ### Do I need special schema or an llms.txt file for AI Overviews? No. Google's documentation says you don't need new machine-readable files, AI text files or markup to appear in AI features, and there is no special schema.org structured data required. ### Can I block AI Overviews but keep ranking in Google? Google says AI is built into Search, so robots.txt rules for Googlebot are the control. To limit what is shown from your pages, use nosnippet, data-nosnippet, max-snippet or noindex. Google-Extended does not affect Search inclusion or ranking. ### How do I see AI Overview traffic in Search Console? Google counts AI Overviews and AI Mode in the overall Performance report under the Web search type; they are not broken out separately in that guidance. ### How often do AI Overviews appear? In Pew's March 2025 sample of 68,879 Google searches by U.S. adults, 18% produced an AI summary. Question-style searches triggered them far more often (60% of queries starting with who, what, when or why). ## Sources 1. [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) 2. [Google — List of Google's common crawlers](https://developers.google.com/search/docs/crawling-indexing/google-common-crawlers) 3. [Pew Research Center — Google users are less likely to click on links when an AI summary appears (22 Jul 2025)](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/) --- # AI visibility tools compared: Otterly, Peec, Profound, Semrush Source: https://citecheck.torontosoft.com/guides/ai-visibility-tools-compared Updated: 2026-11-20 > As of 20 November 2026, published prices: OtterlyAI from €29/month (16 prompts), Peec AI from $95/month (50 prompts, 3 models), Semrush's AI Visibility Toolkit $99/month per domain billed annually (25 prompts), and Profound only offers custom-priced Enterprise plans after retiring its Starter and Growth plans in September 2026. **Disclosure:** CiteCheck is our product. Its free check is listed at the end. Paid CiteCheck plans don't exist yet, so we don't compare them as if they did. Every competitor figure below was read from the vendor's own pricing page on **20 November 2026**. Prices change, so check the linked page before you buy. ## Entry prices and limits | Tool | Entry price (published) | Prompts at entry | Engines / models | Tracking frequency | Source | |---|---|---|---|---|---| | **OtterlyAI** | €29/month (Lite) | 16 | "4 of 7 AI engines" | Daily | [otterly.ai/pricing](https://otterly.ai/pricing) | | **Peec AI** | $95/month (Starter) | 50 | Choose 3 models | Daily | [peec.ai/pricing](https://peec.ai/pricing) | | **Semrush AI Visibility Toolkit** | $99/month per domain, billed annually | 25 | ChatGPT, Google AI, Gemini, Perplexity | Daily | [semrush.com/pricing/ai](https://www.semrush.com/pricing/ai/) | | **Profound** | Custom (Enterprise only) | Custom | Up to 10 answer engines | Daily | [tryprofound.com/pricing](https://www.tryprofound.com/pricing) | ## Higher tiers | Tool | Mid tier | Top published tier | |---|---|---| | OtterlyAI | €189/month, 120 prompts | €489/month, 403 prompts (daily or weekly tracking) | | Peec AI | Pro $245/month, 150 prompts, 2 projects | Advanced $495/month, 350 prompts, 5 projects; Enterprise custom | | Semrush | Add 50 prompts or extra domains as add-ons | — | | Profound | — | Enterprise, custom | ## Tool by tool ### OtterlyAI - **Pricing model:** credits, where one credit is one prompt checked once on one engine. Their FAQ says a prompt tracked daily on one engine uses 31 credits a month, and that Claude answers count as 3 credits because they cost more to analyze. - **Notable:** unlimited brands, competitors, countries and team members on every plan. A free trial is offered. Agency partner program with Looker Studio reporting. - **Good fit:** small teams that want daily tracking at a low entry price. ### Peec AI - **Pricing model:** prompts × models × daily runs (Starter: 50 × 3 × 30 = 4,500 AI answers a month). Extra models are an add-on. - **Notable:** unlimited users. Source and gap analysis, a crawlability audit across 40+ bots, and AI referral tracking. 15% discount on annual billing. - **Good fit:** in-house SEO and content teams that want a deeper analytics layer. ### Semrush AI Visibility Toolkit - **Pricing model:** per domain. $99/month billed annually for 25 daily-tracked prompts and one domain for Brand Performance analysis. - **Notable:** sits next to Semrush's SEO data. Includes prompt research and a site audit for AI readiness. - **Good fit:** teams already paying for Semrush. ### Profound - **Pricing model:** Enterprise only. Profound's FAQ (updated 16 November 2026) says it retired its Starter and Growth plans in September 2026, and the $399 and $499 prices are no longer available for new subscriptions. - **Notable:** tracks up to 10 answer engines, offers agents for marketing tasks, SSO/SAML and SOC 2. - **Good fit:** large brands with an enterprise budget. ## How to choose 1. **Count your prompts first.** Number of questions × markets × engines × frequency. This decides the price more than anything else. 2. **Check which engines you actually need.** For a local service business, Google AI Overviews and ChatGPT may matter more than a long list of models. 3. **Ask how answers are collected.** Official APIs and consumer apps can give different answers. Ask each vendor how they sample, and how they handle the variation between runs. 4. **Look for actions, not just charts.** The value is in knowing which sources to get listed on and which pages to write. ## Where CiteCheck fits today CiteCheck currently offers a **free, no-signup one-off check**: three customer-style questions sent through official AI APIs, with mentions, position, competitors, cited sources, an AI-crawler and structured-data audit of your site, and a fix list. Engine coverage is shown on every result, and engines we haven't connected are marked "not checked" rather than estimated. See [methodology](/methodology). Weekly tracking and agency reports are on the [waitlist](/pricing). ## FAQ ### What is the cheapest AI visibility tracker? Of the four compared here, OtterlyAI's Lite plan at €29/month for 16 prompts had the lowest published entry price on 20 November 2026. Free one-off checks, such as CiteCheck's, cost nothing but don't track over time. ### Does Profound still have a self-serve plan? Profound's pricing FAQ says it retired its Starter and Growth plans for brands in September 2026 and now offers Enterprise plans with custom pricing. ### Is Semrush's AI toolkit monthly or annual? Semrush's AI pricing page shows $99 per month per domain, billed annually, for 25 tracked prompts. ### How many prompts do I need? It depends on how many topics and markets you track. OtterlyAI recommends 50–100 prompts per topic or market for stable trends. ## Sources 1. [OtterlyAI — Pricing](https://otterly.ai/pricing) 2. [Peec AI — Pricing](https://peec.ai/pricing) 3. [Profound — Pricing and FAQ](https://www.tryprofound.com/pricing) 4. [Semrush — AI Visibility Toolkit pricing](https://www.semrush.com/pricing/ai/) --- # GEO checklist: 20 steps to get cited by AI search Source: https://citecheck.torontosoft.com/guides/geo-checklist Updated: 2026-11-20 > GEO (generative engine optimization) means making your business easy for AI assistants to find, understand and cite. The checklist: let AI search crawlers in, keep key content in HTML text, answer real customer questions directly with specific facts and sources, keep business details consistent everywhere, get listed on the sites AI answers already cite, and measure mentions over time. GEO, or **generative engine optimization**, was named in a 2023 research paper from Princeton and IIT Delhi researchers, published at KDD 2024 ([Aggarwal et al.](https://arxiv.org/abs/2311.09735)). Most of the work overlaps with good SEO. The checklist below marks where it differs. ## Access: can AI engines reach you? 1. **Allow AI search crawlers in robots.txt.** At minimum: `OAI-SearchBot` (ChatGPT search), `PerplexityBot`, `Googlebot` (AI Overviews and AI Mode), `Bingbot` and `Claude-SearchBot`. OpenAI says sites opted out of OAI-SearchBot won't be shown in ChatGPT search answers ([OpenAI](https://platform.openai.com/docs/bots)). Test yours with the [AI crawler checker](/tools/ai-crawler-checker). 2. **Decide on training crawlers separately.** `GPTBot`, `ClaudeBot`, `Google-Extended` and `Applebot-Extended` govern training (and, for Google-Extended, Gemini grounding). Blocking them does not remove you from those companies' search features. It's a business choice, not a GEO requirement. See [AI crawlers and robots.txt](/guides/ai-crawlers-robots-txt). 3. **Check your firewall and CDN, not just robots.txt.** Perplexity documents that WAF rules and Cloudflare Bot Fight Mode can block its bots even when robots.txt allows them ([Perplexity](https://docs.perplexity.ai/guides/bots)). Google lists "crawling is allowed … by any CDN or hosting infrastructure" as a best practice ([Google](https://developers.google.com/search/docs/appearance/ai-features)). 4. **Keep important content in HTML text.** If prices, services or locations only appear in images or after JavaScript runs, some crawlers will miss them. 5. **Make pages snippet-eligible.** No accidental `noindex`, `nosnippet` or `max-snippet:0` on pages you want cited. ## Content: are you easy to quote? 6. **Answer first.** Start key pages with a one- or two-sentence direct answer to the question the page targets. 7. **Add specific facts.** Prices or price ranges, service area, hours, years in business, team size, warranty terms. In the GEO study, adding **statistics** was one of the strongest methods ([paper](https://arxiv.org/abs/2311.09735)). 8. **Cite your sources.** Link to the data behind your claims. "Cite Sources" improved visibility most for sources that started low in the ranking. 9. **Use quotations where they're genuine.** Quotes from customers, experts or your own technicians were among the best-performing methods in the study. 10. **Don't keyword-stuff.** The same study found it offered little to no improvement in generative engines. 11. **Write one page per real question.** For example "How much does a roof replacement cost in Denver?" rather than one page trying to rank for everything. 12. **Add comparison tables** where buyers compare options (plans, materials, service tiers). Tables keep facts close to their labels, so they can't be confused with each other. 13. **Keep pages fresh and dated.** Show "last updated" dates on time-sensitive content and actually update it. ## Entity clarity: do engines know who you are? 14. **Use consistent name, address, phone and website** across your site, Google Business Profile, Bing Places and directories. 15. **Add Organization or LocalBusiness structured data** that matches the visible text. Google says no special schema is needed for AI features, but structured data helps search engines understand the page ([structured data guide](/guides/structured-data-for-ai-search)). 16. **Have an About page** with who runs the business, where and since when. ## Off-site: are you where AI answers look? 17. **Find the sources AI answers cite for your category** (run the questions, read the citations) and get listed accurately on them. 18. **Earn reviews on the platforms your customers use.** Review sites often appear as citations in AI answers to local questions. 19. **Contribute genuinely where your buyers ask questions** (industry forums, Reddit communities that allow it). Follow each community's rules on self-promotion. ## Measurement 20. **Track the same questions every week**: mention rate, position, competitors named and cited domains. Also track AI referrals in analytics. ChatGPT adds `utm_source=chatgpt.com` to referral links, according to OpenAI's publisher FAQ ([archived copy](https://web.archive.org/web/20261004224023/https://help.openai.com/en/articles/12627856-publishers-and-developers-faq)). See [How to track AI traffic in Google Analytics](/guides/track-ai-traffic-in-google-analytics). ## Optional extras - **llms.txt.** A proposed file that gives AI agents a curated guide to your site ([llmstxt.org](https://llmstxt.org/)). Google says you don't need it for AI Overviews. It's cheap to add, but don't expect it to change rankings. Use our [llms.txt generator](/tools/llms-txt-generator). ## Start with a baseline Run the [free AI visibility check](/check) before you change anything, so you can measure the effect. ## FAQ ### What is the difference between GEO and SEO? SEO aims to rank pages in search results. GEO aims to have your brand and pages included and cited in AI-generated answers. Most technical SEO still applies; GEO adds attention to direct answers, quotable facts, third-party mentions and measuring AI answers. ### What does the research say works? The GEO paper (Aggarwal et al., KDD 2024) found that adding statistics, quotations and citations improved visibility in generative engine answers by up to 40%, while keyword stuffing did not help. ### How long does GEO take to show results? Crawler changes can be picked up within about a day (OpenAI and Perplexity both cite roughly 24 hours for robots.txt changes). Content and third-party listings take longer, because engines must recrawl and the pages must earn mentions. ## Sources 1. [Aggarwal et al., “GEO: Generative Engine Optimization”, KDD 2024 (arXiv:2311.09735)](https://arxiv.org/abs/2311.09735) 2. [OpenAI — Overview of OpenAI crawlers](https://platform.openai.com/docs/bots) 3. [Perplexity — Perplexity crawlers](https://docs.perplexity.ai/guides/bots) 4. [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) 5. [Anthropic — Does Anthropic crawl data from the web, and how can site owners block the crawler?](https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler) 6. [llmstxt.org — The /llms.txt file, v2 (Jeremy Howard)](https://llmstxt.org/) 7. [OpenAI Help Center — Publishers and Developers FAQ (archived 4 Oct 2026)](https://web.archive.org/web/20261004224023/https://help.openai.com/en/articles/12627856-publishers-and-developers-faq) --- # GEO vs SEO vs AEO: what each term means Source: https://citecheck.torontosoft.com/guides/geo-vs-seo-vs-aeo Updated: 2026-11-20 > SEO optimizes pages to rank in search results. AEO (answer engine optimization) and GEO (generative engine optimization) both mean optimizing to be included and cited in AI-generated answers from ChatGPT, Perplexity, Gemini or Google AI Overviews. GEO comes from a 2023 research paper. In practice the terms overlap, and most of the work builds on SEO. ## Definitions | Term | Stands for | Goal | Main metric | |---|---|---|---| | **SEO** | Search engine optimization | Rank pages in search results | Position, clicks, impressions | | **AEO** | Answer engine optimization | Be the answer an assistant or answer box gives | Mentions and citations in answers | | **GEO** | Generative engine optimization | Be included and cited in AI-generated answers | Visibility/share of voice across prompts | **AEO** and **GEO** are used almost interchangeably. AEO is the older marketing term from the featured-snippet and voice-assistant era. GEO comes from research. ## Where "GEO" came from The term was introduced in **"GEO: Generative Engine Optimization"** by researchers from IIT Delhi and Princeton, among others ([arXiv:2311.09735](https://arxiv.org/abs/2311.09735), published at KDD 2024). The authors built a benchmark of 10,000 queries (GEO-bench) and tested nine ways of rewriting source content to see which ones made it more visible in generated answers. Key findings: - The best methods improved visibility **by up to 40%**. - **Adding statistics, adding quotations and citing sources** performed best. - **Keyword stuffing**, a classic SEO tactic, offered little or no improvement. - Improving **fluency and readability** also helped (15–30% in their tests). - Results **varied by domain**, and the methods also worked when tested on Perplexity.ai. ## What's the same as SEO - Engines that search the web still need to **crawl and index** your pages. Google says pages must be indexed and snippet-eligible to appear in AI Overviews, with "no additional requirements" ([Google](https://developers.google.com/search/docs/appearance/ai-features)). - **Helpful, accurate, people-first content** still wins. - **Technical health** still matters: fast pages, crawlable HTML, internal links and correct structured data. ## What's genuinely different 1. **Being named matters more than being clicked.** Pew found users clicked a traditional result in only 8% of visits when an AI summary appeared, 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/)). If your brand is named in the summary, you still get considered without the click. 2. **Third-party mentions carry more weight.** AI answers often summarise lists, reviews and discussions. Being named on the pages that get cited can matter as much as your own page. 3. **There is no fixed rank.** Answers are generated each time and vary, so you measure a **rate** over many prompts rather than one position. 4. **Separate crawlers.** AI companies use distinct agents for search and training (see [the AI crawler list](/guides/ai-crawlers-robots-txt)). 5. **Quotable facts.** Statistics and specifics give the model something concrete to repeat. ## Which one should a small business focus on? Do the SEO basics first, because they're a precondition. Then add the GEO layer: measure whether AI assistants name you for your top customer questions, and close the gaps. With ChatGPT alone at **1.2 billion weekly users** ([The Verge](https://www.theverge.com/ai-artificial-intelligence/1002265/chatgpt-now-has-1-2-billion-weekly-users-openai-says)), it's worth knowing what it says about your category. Start with the [GEO checklist](/guides/geo-checklist) or a [free AI visibility check](/check). ## FAQ ### Is GEO replacing SEO? No. Google says normal SEO best practices apply to AI Overviews and AI Mode, and AI engines that search the web still need to crawl and index your pages. GEO adds work on top of SEO. ### Where does the term GEO come from? From the paper 'GEO: Generative Engine Optimization' by Aggarwal, Murahari, Rajpurohit, Kalyan, Narasimhan and Deshpande, first posted in November 2023 and published at KDD 2024. ### What is the main metric in GEO? Instead of rank position, GEO looks at whether you are mentioned or cited in the answer, how prominently, and how often across many prompts, often called visibility or share of voice. ## Sources 1. [Aggarwal et al., “GEO: Generative Engine Optimization”, KDD 2024 (arXiv:2311.09735)](https://arxiv.org/abs/2311.09735) 2. [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) 3. [Pew Research Center — Google users are less likely to click on links when an AI summary appears (22 Jul 2025)](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/) 4. [The Verge — ChatGPT now has 1.2 billion weekly users, OpenAI says (29 Sep 2026)](https://www.theverge.com/ai-artificial-intelligence/1002265/chatgpt-now-has-1-2-billion-weekly-users-openai-says) --- # How to check if ChatGPT recommends your business Source: https://citecheck.torontosoft.com/guides/how-to-check-if-chatgpt-recommends-your-business Updated: 2026-11-20 > Ask ChatGPT the same buyer-intent questions your customers ask (for example "best emergency plumber in Toronto"), in a logged-out or fresh chat, several times and on more than one day. Record whether your business is named, its position in the list, which competitors appear and which sources are cited. One answer is a sample, not a ranking. ## Why this matters ChatGPT had **1.2 billion weekly users** at the end of September 2026, according to OpenAI as reported by [The Verge](https://www.theverge.com/ai-artificial-intelligence/1002265/chatgpt-now-has-1-2-billion-weekly-users-openai-says). When a customer asks it "who should I hire for…", the answer is usually a short list. If you're not on it, you don't get considered. Classic rank tracking doesn't show this. A page can rank well on Google and still not be named in an AI answer, and the reverse also happens. ## Step 1: Write the questions your customers actually ask Good test questions have **buying intent** and a **category**, and a **place** if you serve a local area: | Weak question | Better question | |---|---| | "plumber" | "Who is the best emergency plumber in Toronto?" | | "CRM software" | "What's the best CRM for a 5-person roofing company?" | | "dentist" | "Which dentist in Calgary is good for nervous patients?" | Avoid putting your own brand in the question. You want to know whether you come up when the customer doesn't already know you. ## Step 2: Remove personalisation OpenAI says ChatGPT search can rewrite your question into search queries using your **general location from your IP address**, and, if memory is turned on, **your saved memories** ([OpenAI Help Center](https://help.openai.com/en/articles/9237897-searching-the-web-with-chatgpt)). If you test from your own office while logged in, the answer may be tailored to you. - Use a new chat. Turn memory off, or test while logged out. - If you serve another city, put the city in the question rather than relying on location. ## Step 3: Record the same four things every time 1. **Mentioned?** Is your business named anywhere in the answer? 2. **Position.** If the answer is a list, where are you in it? 3. **Competitors.** Who is named instead of you, or ahead of you? 4. **Sources.** Which websites are cited or linked? These are often directories, review sites, Reddit threads or "best of" articles. A simple spreadsheet works: one row per question and run, with columns for each of the four. ## Step 4: Repeat, because answers vary AI answers are generated fresh each time, so the same question can produce different lists. Vendors that track AI visibility for a living say this too: OtterlyAI recommends **50 to 100 prompts per topic** and says tracking 100 prompts daily for 30 days reduced variation in brand coverage by about 73% compared with tracking 10 ([OtterlyAI pricing FAQ](https://otterly.ai/pricing)). For a small business, a practical minimum is 3–10 questions, asked 3 times each, once a week. The number that matters is the **share of answers that name you**, and whether it goes up. ## Step 5: Check that ChatGPT can reach your site at all ChatGPT's search features rely on OpenAI's crawler **OAI-SearchBot**. OpenAI says sites that opt out of OAI-SearchBot will not be shown in ChatGPT search answers, though they can still appear as navigational links ([OpenAI crawler docs](https://platform.openai.com/docs/bots)). Blocking **GPTBot** is a separate choice: it only signals that your content shouldn't be used for training. Check your robots.txt with our free [AI crawler checker](/tools/ai-crawler-checker). ## Faster: run the free check The [free CiteCheck checker](/check) automates steps 1–3: it builds three buyer-intent questions from your category and city (or uses your own), sends them to AI engines through their official APIs, and shows whether you're mentioned, your position, the competitors named and the sources cited, plus a fix list based on what it found. It also checks your robots.txt for AI crawlers and your homepage's structured data. It's an API-based sample, so treat it the way you'd treat your own spreadsheet: a consistent measurement to compare over time, not a guarantee of what every user sees. See [how the checks work](/methodology). ## What to do when you're not mentioned - **Look at the sources.** If the answer cites a directory, a review site or a listicle that names your competitors, getting accurately listed there is usually the most direct fix. - **Answer the question on your own site.** A clear page that states what you do, where, for whom and at what price is easier to cite than a vague homepage. - **Keep your details consistent** across your site, Google Business Profile and directories. Our [GEO checklist](/guides/geo-checklist) covers the full list. ## FAQ ### Does ChatGPT give everyone the same recommendations? No. Answers vary between runs, and ChatGPT can use your approximate location (from your IP address) and, if memory is enabled, saved memories when it rewrites a search query, according to OpenAI's help center. Test several times and from a clean chat. ### Can I check this for free? Yes. You can test manually in ChatGPT, or use the free CiteCheck checker, which sends three customer-style questions to AI engines through their official APIs and reports mentions, positions, competitors and cited sources. ### Are API results the same as what I see in the ChatGPT app? Not exactly. The consumer app adds personalisation, location and product features that APIs don't. API checks are best used as a consistent, repeatable sample to compare over time and against competitors. ### How many questions should I test? Start with 3 to 10 questions that real customers ask before buying. Vendors that track this at scale recommend far more: OtterlyAI suggests 50 to 100 prompts per topic for stable trends. ## Sources 1. [OpenAI Help Center — Searching the web with ChatGPT](https://help.openai.com/en/articles/9237897-searching-the-web-with-chatgpt) 2. [OpenAI — Overview of OpenAI crawlers](https://platform.openai.com/docs/bots) 3. [The Verge — ChatGPT now has 1.2 billion weekly users, OpenAI says (29 Sep 2026)](https://www.theverge.com/ai-artificial-intelligence/1002265/chatgpt-now-has-1-2-billion-weekly-users-openai-says) 4. [OtterlyAI — Pricing](https://otterly.ai/pricing) 5. [Pew Research Center — Google users are less likely to click on links when an AI summary appears (22 Jul 2025)](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/) --- # How to get mentioned in Perplexity answers Source: https://citecheck.torontosoft.com/guides/how-to-get-mentioned-in-perplexity Updated: 2026-11-20 > Perplexity builds answers from web pages it can retrieve and links them as citations. To be mentioned: allow PerplexityBot in robots.txt and in your firewall, publish pages that answer specific questions with concrete facts, and get listed on the third-party pages Perplexity already cites for your category. ## How Perplexity decides what to mention Perplexity answers questions by retrieving web pages and summarising them, with numbered citations linking to the pages it used. That has a practical consequence: **if Perplexity can't fetch or find your page, it can't cite it**, and the businesses it names are usually the ones that appear on the pages it retrieved. So "getting mentioned" comes down to two jobs: 1. Make your own pages **reachable and quotable**. 2. Be present on the **third-party pages** Perplexity already uses for your category. ## 1. Let Perplexity's crawler in Perplexity documents two user agents ([Perplexity crawlers](https://docs.perplexity.ai/guides/bots)): | User agent | What it does | Respects robots.txt? | |---|---|---| | **PerplexityBot** | Surfaces and links websites in Perplexity search results. Not used to crawl content for AI foundation models. | Yes | | **Perplexity-User** | Visits a page when a user asks a question, to answer it and link the page. | Generally no, because a user requested the fetch | Perplexity recommends **allowing PerplexityBot** in robots.txt and allowing its published IP ranges. Changes can take up to 24 hours to take effect. A minimal robots.txt rule: ``` User-agent: PerplexityBot Allow: / ``` If you use **Cloudflare or AWS WAF**, a bot rule can block Perplexity even when robots.txt allows it. Perplexity's docs explain how to create a skip rule that matches both the user agent and its IP ranges, and warn that Cloudflare's Bot Fight Mode can't be skipped with custom rules. Test your current setup with the free [AI crawler checker](/tools/ai-crawler-checker). ## 2. Publish pages that answer one question well Research on "generative engine optimization" (GEO) tested what makes a source more visible in AI-generated answers. The paper by Aggarwal et al. ([KDD 2024](https://arxiv.org/abs/2311.09735)) found: - Adding **statistics**, **quotations** and **citations of sources** to content increased its visibility in generative engine answers, by up to 40% in their benchmark. - **Keyword stuffing** offered little to no improvement. - The methods were also tested on **Perplexity.ai** itself, not only on a simulated engine. - Effects varied by topic, and lower-ranked sources gained the most from citing sources. In practice, for a business page: - Put the **direct answer first**: what you do, for whom, where, and at what price range. - Add **specific facts** a summary can quote: years in business, number of jobs or clients, service area, certifications, warranty terms. - **Cite your sources** for any claims about your market or results. - Use **headings that match questions** ("How much does a water heater replacement cost in Toronto?"). ## 3. Be where Perplexity already looks Run the questions your customers ask and look at the **citations** under each answer. For local services and product categories they are often directories, review platforms, Reddit threads, YouTube videos and "best of" articles. Pew Research found that Wikipedia, YouTube and Reddit were the most frequently cited sources in Google's AI summaries, accounting for 15% of the sources listed ([Pew Research Center](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/)). Perplexity's mix differs by query, so check your own category. For each cited page that names competitors but not you, ask: can I be listed there accurately? Directories and review sites usually have a submission or claim process. Listicles have authors you can contact with useful, verifiable information. ## 4. Measure, then repeat Use the [free checker](/check) or the manual method in [How to check if ChatGPT recommends your business](/guides/how-to-check-if-chatgpt-recommends-your-business) for the same questions every week. You're looking for two numbers to rise: the **share of answers that name you**, and how often **your own domain is cited**. ## FAQ ### What is PerplexityBot? PerplexityBot is Perplexity's crawler for surfacing and linking websites in its search results. Perplexity says it is not used to crawl content for AI foundation models and recommends allowing it in robots.txt. ### Does blocking Perplexity-User stop Perplexity from reading my page? Not reliably. Perplexity says Perplexity-User fetches pages when a user asks a question and, because a user requested the fetch, it generally ignores robots.txt rules. ### My firewall blocks bots. Is that a problem? It can be. Perplexity documents how to allow its bots in Cloudflare and AWS WAF, and notes that Cloudflare's Bot Fight Mode can still challenge them even with a skip rule. ### Does adding statistics or quotes help? In the GEO research paper (KDD 2024), adding statistics, quotations and source citations to content increased its visibility in generative engine answers, including in tests on Perplexity.ai. Results varied by domain. ## Sources 1. [Perplexity — Perplexity crawlers](https://docs.perplexity.ai/guides/bots) 2. [Aggarwal et al., “GEO: Generative Engine Optimization”, KDD 2024 (arXiv:2311.09735)](https://arxiv.org/abs/2311.09735) 3. [Pew Research Center — Google users are less likely to click on links when an AI summary appears (22 Jul 2025)](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/) --- # How AI assistants choose which local businesses to recommend Source: https://citecheck.torontosoft.com/guides/local-business-ai-recommendations Updated: 2026-11-20 > AI assistants that search the web turn a local question into search queries, often including your city or the user's approximate location, then summarise what the results say. The businesses named are usually the ones that appear consistently on well-cited pages: maps listings, review sites, directories and local articles. Accurate listings and reviews, plus a site that states clearly what you do and where, are the main levers. ## What the platforms document **ChatGPT.** OpenAI says ChatGPT search "sometimes partners with other search providers" and **rewrites your question into one or more targeted search queries**. It collects **general location from your IP address**, so "good restaurants near me" can become "top restaurants San Francisco". If memory is on, saved memories can shape the query too. OpenAI also says ChatGPT ranks results using multiple factors and that **placement is not guaranteed** ([OpenAI Help Center](https://help.openai.com/en/articles/9237897-searching-the-web-with-chatgpt)). **Google AI Overviews and AI Mode.** Google says the same SEO fundamentals apply, and lists **"checking that your Merchant Center and Business Profile information is up-to-date"** among its best practices ([Google Search Central](https://developers.google.com/search/docs/appearance/ai-features)). Nobody publishes a full ranking formula. But one consequence of the documented process is clear: **an assistant can only name businesses that appear in what it retrieves.** ## The signals you control | Signal | Why it matters | What to do | |---|---|---| | **Listings** (Google Business Profile, Bing Places, Apple Business Connect, industry directories) | Local search results and directories are what gets retrieved for "best X in Y" queries | Claim them, complete every field, use the same name, address and phone everywhere | | **Reviews** | Review sites and review counts are commonly summarised in answers | Ask every happy customer, reply to reviews, stay on the platforms customers in your trade use | | **"Best of" and local articles** | Listicles are frequently cited for recommendation questions | Find the ones cited for your category and city, and pitch accurate, useful information to their authors | | **Your own site** | Gets cited when it answers the question directly | State services, area, prices and credentials in plain text, with one page per main service | | **Structured data** | Removes ambiguity about who and where you are | Add LocalBusiness JSON-LD that matches the page ([guide](/guides/structured-data-for-ai-search)) | | **Crawler access** | Blocked sites can't be retrieved or cited | Allow OAI-SearchBot, PerplexityBot, Googlebot and Bingbot ([check yours](/tools/ai-crawler-checker)) | ## A four-week plan **Week 1: baseline.** Write five questions customers ask ("best [service] in [city]", "emergency [service] [city] open now", "[service] near [neighbourhood]"). Run a [free check](/check) and note who gets named and which sources are cited. **Week 2: listings.** Fix your Google Business Profile, Bing Places and the top three directories cited in week 1. Make name, address, phone, hours and categories identical everywhere. **Week 3: on-site answers.** Make sure your homepage says in the first lines what you do and where. Add or improve one page per core service, with prices or ranges, service area, response times and credentials. Add LocalBusiness markup. **Week 4: reputation.** Ask recent customers for reviews. Contact authors of cited local "best of" lists with factual information (licence, years, specialties). Re-run the same questions and compare. ## Why results differ between people Because location and memory feed into the search query, two people asking the same question can see different businesses. For your own testing, **put the city in the question** and use a fresh chat. Look at the **share of answers** that name you over several runs, not one screenshot. ## Where AI answers get their sources Pew Research found that Wikipedia, YouTube and Reddit were the most frequently cited sources in Google's AI summaries in its March 2025 study, together making up 15% of the sources listed ([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/)). For local services, the citations you'll see are usually more specific: maps, review platforms and local directories. Run your own questions and read the citations. ## FAQ ### Does ChatGPT know where the user is? OpenAI says ChatGPT collects general location from the user's IP address and may share that general location with third-party search providers, for example rewriting 'good restaurants near me' as 'top restaurants San Francisco'. ### Does my Google Business Profile matter for AI? For Google's AI features, yes: Google lists keeping your Business Profile and Merchant Center information up to date among the best practices for AI Overviews and AI Mode. ### Can I pay to be recommended by ChatGPT? There's no documented way to pay for placement in ChatGPT's organic recommendations. OpenAI says placement in search results is not guaranteed and that ChatGPT ranks results using multiple factors intended to help users find relevant, reliable information. ## Sources 1. [OpenAI Help Center — Searching the web with ChatGPT](https://help.openai.com/en/articles/9237897-searching-the-web-with-chatgpt) 2. [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) 3. [Pew Research Center — Google users are less likely to click on links when an AI summary appears (22 Jul 2025)](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/) 4. [Google Search Central — Local business structured data](https://developers.google.com/search/docs/appearance/structured-data/local-business) --- # Structured data for AI search: what helps and what doesn't Source: https://citecheck.torontosoft.com/guides/structured-data-for-ai-search Updated: 2026-11-20 > Google says no special schema.org markup is needed to appear in AI Overviews or AI Mode. Structured data still helps search engines understand your page, as long as it matches the visible text. For most businesses the useful core is Organization or LocalBusiness, plus Product, Article or BreadcrumbList where they fit. ## What Google actually says From Google's AI features documentation ([Google Search Central](https://developers.google.com/search/docs/appearance/ai-features)): - "There's also no special schema.org structured data that you need to add" to appear in AI Overviews or AI Mode. - One of its listed best practices is "making sure your structured data matches the visible text on the page". From Google's introduction to structured data ([Google](https://developers.google.com/search/docs/appearance/structured-data/intro-structured-data)): structured data gives **explicit clues about the meaning of a page**, and Google recommends **JSON-LD** as the format. So structured data isn't a shortcut into AI answers. It's a way to state facts about your business unambiguously, which any system that reads it can use. ## The types worth adding | Type | Use it for | Notes | |---|---|---| | `Organization` | Every business site, homepage or About page | Name, URL, logo, `sameAs` links to your official profiles | | `LocalBusiness` (or a subtype such as `Plumber` or `Dentist`) | Businesses with a physical location or service area | Address, phone, opening hours, geo coordinates ([Google's guide](https://developers.google.com/search/docs/appearance/structured-data/local-business)) | | `Product` / `Offer` | Product and pricing pages | Price, availability, reviews if shown | | `Article` / `BlogPosting` | Guides and posts | Headline, dates, author/publisher | | `BreadcrumbList` | Any page deeper than the homepage | Shows the page's place in your site | | `FAQPage` | Visible Q&A sections | Google rich results limited to government and health sites since 2023 | ## A LocalBusiness example ```json { "@context": "https://schema.org", "@type": "Plumber", "name": "Northside Plumbing", "url": "https://www.example.com/", "telephone": "+1-416-555-0100", "address": { "@type": "PostalAddress", "streetAddress": "100 Example St", "addressLocality": "Toronto", "addressRegion": "ON", "postalCode": "M5V 1A1", "addressCountry": "CA" }, "areaServed": ["Toronto", "North York", "Scarborough"], "openingHoursSpecification": [{ "@type": "OpeningHoursSpecification", "dayOfWeek": ["Monday","Tuesday","Wednesday","Thursday","Friday","Saturday","Sunday"], "opens": "00:00", "closes": "23:59" }], "sameAs": ["https://www.facebook.com/example", "https://www.yelp.com/biz/example"] } ``` Put it in a `