← Insights
FAQ

GEO FAQ: the questions B2B marketing teams ask about AI search

The real questions buyers ask AI engines about GEO, each with a short, sourced answer and a link to the page that goes deeper.

·25 min read·By Jonathan Taieb, Founder and CEO
Key takeaways
  • GEO is the work of getting your brand named, and your pages cited, in the answers of ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews.
  • AI answers change from one run to the next: reliable measurement asks the same questions several times and compares changes to a noise threshold.
  • Much of what gets cited lives off your site: comparisons, review sites, trade media, YouTube, Reddit.
  • No agency can guarantee placement in ChatGPT: OpenAI itself says placement is not guaranteed.

Understanding GEO

What GEO is, how AI engines pick their answers, and why those answers change.

What is generative engine optimization and does my business need it?

Generative engine optimization (GEO) is the work of getting your brand named, and your pages cited, in the answers that ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews write when buyers ask their questions. The term comes from a 2023 research paper [1]. You need it if your buyers ask AI assistants for vendor shortlists: in G2’s 2026 survey, 51% of B2B software buyers said they start research with an AI assistant more often than with Google [2].

Go further: The complete GEO guide

GEO vs SEO: what’s the difference and where should I invest?

SEO ranks a page in a list of links; it works keyword by keyword and is measured in positions and clicks. GEO puts your brand inside the generated answer; it works question by question and is measured in the share of answers that name you, engine by engine. Invest in both, in that order: crawlable, indexed, well-structured pages are the foundation GEO builds on.

Go further: GEO vs SEO for B2B buyers

Is SEO dead because of AI search?

No. AI engines retrieve live web pages for many questions, and Google says there are no extra requirements to appear in AI Overviews beyond being indexed and eligible for a snippet [4]. What changes is the click: in Pew’s US panel, users clicked a traditional result on 8% of visits when an AI summary appeared, against 15% without one [3]. SEO remains the foundation; the answer itself becomes a second goal.

Go further: GEO vs SEO for B2B buyers

How does ChatGPT decide which brands to recommend?

With search on, ChatGPT rewrites the question into one or more targeted searches, retrieves results from third-party search providers and partners, and ranks them on multiple factors; OpenAI says placement is not guaranteed [6]. Brands that appear on the pages it retrieves, with consistent descriptions across several sources, are easier to recommend. Without search, it answers from what the model learned in training.

Go further: How AI engines choose brands

How do I get my company to show up in ChatGPT, Claude, and Gemini answers?

Three conditions, in order. Be readable: allow each engine’s search crawler, OAI-SearchBot for ChatGPT, Claude-SearchBot for Claude, Googlebot for Gemini and AI Overviews [4][5][15]. Be relevant: publish a page that answers each buying question precisely. Be corroborated: get described consistently on the third-party sites these engines reuse. Each engine reads different sources, so measure them separately.

Go further: Measuring visibility engine by engine

Why does my site rank on Google but never show up in ChatGPT answers?

The usual causes: OAI-SearchBot blocked in robots.txt or by a firewall rule [5]; a site poorly known to Bing or the other search providers ChatGPT draws on [6]; pages that rank for keywords but answer no specific buying question; and a brand missing from the third-party lists and comparisons ChatGPT reuses. A good Google ranking helps, but it is not enough on its own.

Go further: Google yes, ChatGPT no: the causes

Why does ChatGPT recommend my competitors instead of me?

Because your competitors are present where ChatGPT looks: pages that answer the exact question, and third-party sources (comparisons, lists, reviews, media) that mention them. In one study of B2B software recommendations, the recommended tool’s own site was cited only 11.6% of the time [20]: recommendations are won largely off your own site. Measure who gets named instead of you, and from which sources.

Go further: Why competitors get named

ChatGPT gives different answers every time, so how do you measure AI visibility reliably?

By repeating the measurement. In a SparkToro study, ChatGPT and Google’s AI almost never returned the same list of brands twice across 100 runs [8]. A reliable protocol asks a fixed set of buying questions several times on each engine, 5 times at ZivRank, counts “named” and “cited” separately, and compares any change to a noise threshold set in advance. A single screenshot proves nothing.

Go further: Our measurement protocol

How long does it take to start appearing in ChatGPT answers?

Expect weeks, rarely less. A page has to be crawled, indexed by the search providers ChatGPT uses, and then picked up in answers. Across ZivRank programs, the observed pickup time is 3 to 8 weeks after publication. A mention on a third-party page that ChatGPT already cites can work faster than a new page on your own site.

Go further: The GEO guide

Getting cited: content and sources

Which content AI engines reuse, where to publish it, and what not to do.

What kind of content gets cited by ChatGPT and Claude in B2B?

Pages that answer a buying question directly, structured with lists and tables. In a study of 40 software categories, every page ChatGPT cited was list-structured, and most had a year in the title and a comparison table [20]. The original GEO research also found that adding citations, statistics and quotations improved visibility, while keyword stuffing did not [1].

Go further: How to get cited by ChatGPT

How do I write content that LLMs will quote and cite?

One question per page, asked in the title. The answer in the first lines, in two to four self-contained sentences. Then question-style subheadings, lists, a table when you compare, dated figures with a linked source. A named author and an update date. Text that renders without JavaScript. Avoid vague claims: an AI quotes a precise sentence, not a slogan.

Go further: Our production plan

Should B2B companies publish "X vs competitor" comparison pages to win AI search?

Yes, if they are honest. Buyers ask AI engines “X vs Y” and “alternatives to X”, and comparison pages with tables are a format these engines reuse [20]. Keep them factual: compare on criteria you can source, date the page, link to the competitor’s own pages, and never disparage. A one-sided page is easy to spot, for buyers and for engines that cross-check other sources.

Go further: Our comparison pages

How do I get my company listed in the "best of" listicles that AI tools cite?

First find which lists the engines actually cite on your buying questions: ask them, several times, and note the source links. Then, for each list: contact the author with verifiable facts (who you serve, pricing, integrations, customers you can name), keep your review profiles current, and offer a demo or data. Prioritize lists cited by several engines across many questions.

Go further: The sources AI engines reuse

How do I get my brand mentioned on the sites AI engines use as sources, like G2, review sites and trade publications?

Measure which sources the engines cite on your questions; they differ by market. Then work each one: complete and update your review-site profiles and ask happy customers for reviews, offer original data to trade publications, and ask to be included in the comparisons and lists already cited, with verifiable facts. A source cited by several engines across many questions comes first.

Go further: G2, Capterra and AI citations

Do YouTube videos help brands get cited by AI assistants?

Often, yes: YouTube is among the domains AI engines cite most, and it was the top cited source on a B2B market we measured in August 2026. A useful video answers one buying question, uses that question in its title and chapters, and publishes a transcript. Link it to the page on your site that develops the answer.

Go further: The sources AI engines reuse

Does LinkedIn content influence how ChatGPT recommends B2B vendors?

Somewhat, but not the way most people think. What we observe: engines reuse long, structured articles published on LinkedIn and expert profiles whose expertise is documented elsewhere, more than short posts. A viral post that no indexed source picks up carries little weight. Publish long, signed articles that answer buying questions with data, and get them picked up by trade media.

Go further: How AI engines choose brands

Does Reddit influence ChatGPT recommendations, and how can brands show up there safely?

Reddit is among the domains AI engines cite most, Perplexity in particular according to several studies. To show up safely: use a named account that discloses its link to the company, answer questions that are already being asked, avoid vote campaigns and fake accounts, and follow each community’s rules. Reddit’s rules prohibit manipulation and spam [16].

Go further: Reddit and AI answers in B2B

How can a mid-market company get recommended by ChatGPT?

By being specific. A mid-market company rarely wins “best software”, but it can win “best tool for this team, this industry, this company size or this compliance need”. Pick a few dozen questions like that, publish a page that answers each one, keep your review profiles and integration listings current, and check that the engines’ crawlers can read your site.

Go further: The mid-market playbook

How do I fix wrong information about my company in ChatGPT?

Fix the sources first. Find, in the answers or their links, the page the error comes from. Update your own site with the correct facts, stated plainly, then your third-party profiles: directories, review sites, partner pages, and Wikipedia if a page exists. Report the wrong answer with ChatGPT’s feedback button. The correction shows once the engines have re-read those sources, usually after several weeks.

Go further: Tracing the sources behind answers

Can you manipulate ChatGPT answers? What are the risks?

People try: hidden text aimed at AI, fake reviews, fake forum accounts, pages produced at scale. The risks are real. In the US, the FTC rule on consumer reviews bans fake reviews and testimonials [17]. Google penalizes scaled content with no value [12]. Reddit bans manipulation [16]. And a brand caught out loses buyers’ trust. Durable GEO means being the best answer, not faking it.

Go further: Our method commitments

How do I get my brand cited in Google AI Overviews?

Google says there is no special requirement for AI Overviews beyond being indexed and eligible to be shown with a snippet [4]. So the work is solid SEO plus short, precise answers near the top of the page, under question-style headings. AI Overviews and AI Mode cite different pages for the same query [19]: track them separately, over several runs.

Go further: AI Mode vs AI Overviews

How can a European company build AI visibility in the US market?

Measure the US market on its own, in English, with your US buyers’ questions: the sources cited there are not the ones cited in Europe. Publish native English pages, not translations, and make sure your US pages are indexed by both Google and Bing. Then work US sources: review sites, US trade publications and the lists that rank for your category.

Go further: Building AI visibility in the US

Measuring AI visibility

What to count, with which tools, and how to connect visibility to pipeline.

How can I tell if ChatGPT mentions my brand when buyers ask for vendor recommendations?

Ask ChatGPT the questions your buyers ask, with search on, several times each, and record whether your brand is named, cited as a source, or absent. One run proves nothing because answers vary [8]. At scale, that means a fixed question set, repeated runs and a log of every answer, which is what tracking tools and measurement protocols automate.

Go further: Free Quick Scan

How can I find out which sources ChatGPT cites in my industry?

Ask ChatGPT, with search on, the questions your buyers ask, several times each, and log the links under each answer. Count two things per domain: total citations, and the number of different questions it appears on. A domain cited across many questions is part of the engine’s core corpus for your category: that is where you need to be first.

Go further: Our source analysis

How do I track my visibility in Google AI Mode?

Separately from AI Overviews. The two surfaces can use different models and cite different pages: Ahrefs found only 13.7% of cited URLs in common for the same query [19]. Search Console adds their traffic into the overall Web report [4]. To know where you are cited, run your buying questions in AI Mode several times and log the brands and sources shown.

Go further: AI Mode vs AI Overviews

What tool tracks brand mentions in ChatGPT and Claude over time?

Two families. Tracking software (Peec AI, Profound, Otterly.AI, Meteoria, Qwairy, among others) often measures daily, self-serve. Agencies measure and also produce the content. ZivRank measures up to 5 engines (ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews) every month with 5 runs per question; its Scan Continu plan, on 3 engines, starts at $390 per month.

Go further: ZivRank plans

What should an AI visibility dashboard include?

Five numbers, each with its definition: named rate (answers naming you over all answers in the protocol), cited rate (answers citing one of your pages), share of voice against a competitor list you validated, your site’s rank among cited domains, and AI referral traffic. Plus what makes them checkable: the question set, engines, runs per question, margin of error and the noise threshold.

Go further: GEO KPIs to track

How do I benchmark my AI visibility against competitors?

Share of voice: your mentions divided by your mentions plus your competitors’, on the same questions, engines and number of runs. Two choices drive the result: the competitor list, which you should validate yourself rather than leave to an algorithm, and any weighting, for example by search volume [21]. Get both in writing, or the number is not comparable from month to month.

Go further: GEO KPIs to track

What methodology should a GEO agency use to measure results?

Five rules. A fixed set of buying questions, agreed before measuring. Several runs per question and engine, because answers vary [8]. Separate counts for named, cited and absent. Questions that already contain your brand set apart. A noise threshold measured by running the same protocol twice: at ZivRank, any change within ±15% is not reported as a result. And the raw answers handed over.

Go further: The ZivRank protocol

How do I run an AI visibility audit for my company?

Four steps. List your buying questions, from generic to precise. Ask them on each AI engine, several times, and record whether you are named, cited or absent, and who is named instead. Log the cited domains to see where you need to be present. Then check your site: AI crawler access, text readable without JavaScript, answers near the top of pages, structured data.

Go further: The audit, step by step

How do I attribute pipeline and leads to ChatGPT and Claude mentions?

Combine three signals. Traffic: GA4’s AI Assistant channel, live since May 13, 2026, and ChatGPT’s utm_source=chatgpt.com parameter [7][9]. Self-reported source: a “how did you hear about us?” field with a “ChatGPT or another AI assistant” option, carried into your CRM. Correlation: compare those leads with your citation rate over time. None is enough alone; together they give a defensible order of magnitude.

Go further: GEO KPIs to track

Is there a free AI visibility check for my brand?

Yes. Several tracking tools offer a free self-serve trial. The most useful test uses your own buying questions, not generic ones, on several engines and several runs. At ZivRank, the Quick Scan is free: up to 5 of your buying questions, with a video walkthrough by an expert within 48 hours.

Go further: Request the free Quick Scan

Are there GEO case studies with measured B2B results?

A few, and they are worth reading for their method as much as their numbers. Our published case is a US B2B SaaS vendor measured twice with the same protocol, 44 days apart: answers naming the brand on Perplexity went from 21 to 50, while ChatGPT stayed at 17. Ask any provider for both measurements, not just the result.

Go further: Reading GEO case studies

Tracking tools: Profound, Peec AI, Otterly.AI and others

What the main tracking tools do, based on their own pages. Our comparison pages go into each one.

AI visibility tool or GEO agency: which should I choose?

A tracking tool tells you where you stand; it does not produce the content for you. Choose a tool if your team has time to write, publish and distribute. Choose an agency if you lack hands, or want a third party to prove the effect of its own work. They combine well: many teams keep their tool and hand off the execution.

Go further: Tool or agency: how to decide

Profound vs Peec AI: which AI visibility tool is better?

Neither is better in general; they fit different teams. Profound is a software platform for tracking visibility in answer engines, with an "AI Marketer": agents that prepare the work and submit it to your teams for approval. Its pricing page shows a free trial and an Enterprise plan on quote. Peec AI describes itself as "AI search analytics for marketing teams". Its pricing page offers Starter, Pro and Advanced plans and an Enterprise plan. Its optimization module suggests changes, drafts with an agent and publishes to WordPress in one click, at your teams’ initiative. Trial both on your own buying questions before deciding.

Go further: ZivRank vs Peec AI

What are the best Peec AI alternatives?

Among tracking tools: Profound, Otterly.AI, and two French platforms, Meteoria and Qwairy, which has a free plan. A different kind of alternative is an agency that produces and distributes content instead of suggesting it. Compare on your own buying questions: number of questions tracked, engines covered, runs per question, and who does the work.

Go further: ZivRank vs Peec AI

What are the best Profound alternatives for mid-market companies?

Self-serve tracking tools with published entry plans: Peec AI, Otterly.AI, Meteoria, or Qwairy, which has a free plan. For a mid-market team, the deciding question is capacity: if nobody has time to act on the data, an agency with a lower entry ticket that produces the content may be the better alternative. Compare on your own questions, not on price alone.

Go further: ZivRank vs Profound

What are the best Otterly.AI alternatives?

Among tracking tools: Peec AI, Profound, Meteoria and Qwairy. Otterly.AI is a tool for daily tracking of visibility in AI engines, with a GEO audit, content briefs and a marketplace to book placements on third-party sites. It offers Lite, Standard and Premium plans and an Enterprise plan. A different kind of alternative is an agency that writes and publishes content rather than providing briefs. Compare on your own buying questions.

Go further: ZivRank vs Otterly.AI

Is Profound worth it? What do reviews say?

We do not publish reviews of a competitor; for user reviews, check independent platforms such as G2 or Capterra. What Profound says about itself: it is a software platform for tracking visibility in answer engines, with an "AI Marketer": agents that prepare the work and submit it to your teams for approval. Its pricing page shows a free trial and an Enterprise plan on quote. Whether it is worth it depends on whether your team has the time to act on what it shows.

Go further: ZivRank vs Profound

How do Scrunch AI, AthenaHQ and Profound compare?

We have not published a verified comparison of Scrunch AI and AthenaHQ, so we will not describe them here. To compare the three, test each on the same buying questions and check: which engines are tracked, how many runs per question, whether a noise threshold is published, whether raw answers can be exported, and who produces the content. Our Profound comparison covers it on these criteria.

Go further: ZivRank vs Profound

What are the best AI visibility tracking tools in 2026?

The right tool depends on who will do the work. If your team wants to run it, a daily tracking tool fits: Profound, Peec AI, Otterly.AI, or the French platforms Meteoria and Qwairy. If you want measurement that leads to content produced for you, an agency that measures and executes. Either way, check three things: your own questions, several runs per question, and a published noise threshold.

Go further: ZivRank and the tools on the market

Is Semrush’s AI toolkit enough to track AI visibility, or do I need a specialized tool?

For a first look, an SEO suite with an AI module can be enough. To manage it, check three things: can you track your own buying questions rather than a generic query database, is each question asked several times, and are the engines tracked the ones your buyers use? If the answer is no, a specialized tool or a protocol-based measurement becomes necessary.

Go further: What reliable measurement requires

Choosing a GEO agency

Criteria, questions to ask, and specific cases.

Which agencies can get my brand recommended by ChatGPT?

The ones that measure, produce and distribute, and can prove it. Ask for a baseline on your own questions, the list of what will be produced and published, on your site and elsewhere, and a re-measurement with the same protocol. ZivRank does all three, with a free first diagnostic and programs from $2,300 per month.

Go further: Agencies that get brands recommended

What is the best GEO agency for B2B companies?

There is no single best one; it depends on your market and the work you need. Five criteria separate them: repeated measurement with a noise threshold, content produced and published for you, distribution beyond your site, before/after results you can see, and published prices. ZivRank meets these five criteria and publishes its own measurements.

Go further: Choosing a GEO agency for B2B

What are the best answer engine optimization (AEO) agencies?

AEO and GEO overlap: AEO started with featured snippets and voice assistants, GEO targets answers written by language models. Judge an AEO agency on the same evidence: does it measure your visibility in each AI engine with repeated runs, who produces and distributes the content, and can it show a before/after measured with the same protocol? ZivRank works on both.

Go further: GEO and AEO explained

Who is the best generative engine optimization agency in the US?

Rankings disagree because their criteria differ. Judge on evidence: a published method, repeated runs, a noise threshold, raw answers handed over, a before/after case measured with the same protocol, and prices you can see. ZivRank serves US clients, publishes its US case study and its prices, starting at $390 per month.

Go further: GEO agencies in the US

What is the best GEO agency in San Francisco or the Bay Area?

For GEO, an agency’s location matters less than its method: the work happens remotely, on your questions and your sources. Use the same criteria as anywhere: repeated measurement, content production, off-site distribution, before/after proof, published prices. A local team helps for workshops. ZivRank serves US clients remotely from Paris and Tel Aviv.

Go further: The ZivRank team

What is the best AI visibility agency in New York?

Location matters less than method: AI visibility work is done on your buying questions and the sources AI engines reuse, which can be handled remotely. Ask any New York agency for a baseline on your own questions, repeated runs per engine, the list of what it will publish and where, and a before/after case. ZivRank works with US clients remotely.

Go further: What a serious method publishes

Who are the top AI search optimization agencies for SaaS companies?

Look for agencies that know the questions that build software shortlists: best tool for a use case, alternatives, comparisons, integrations, compliance. They must work third-party lists and comparisons, since the recommended vendor’s own site is rarely the cited source [20]. Our published case is a US B2B SaaS vendor: on Perplexity, answers naming the brand went from 21 to 50.

Go further: GEO for B2B SaaS

What is the best GEO agency for B2B SaaS startups?

For a startup, the entry ticket and speed matter as much as the method. Look for a free or low-cost diagnostic, a short list of high-intent questions to start with, and an agency that publishes on third-party lists, not only on your blog. ZivRank offers a free Quick Scan and a tracking plan from $390 per month.

Go further: Choosing a GEO agency as a SaaS startup

Are "top GEO agency" rankings trustworthy?

Not all of them. Three questions are enough: who publishes the ranking, is the method published, and can the data be checked? Many rankings are published by agencies themselves. ZivRank publishes some too, with its method and sources: ask it the same questions. A ranking based on repeated measurement in AI engines is worth more than an editorial list.

Go further: Our method

Should my SEO agency handle GEO, or do I need a specialist?

Your SEO agency can do GEO if it can answer three questions: how does it measure your visibility in each AI engine, with how many runs per question; who produces and distributes content beyond your site; and which before/after results can it show? If the answers are vague, it is rebranded SEO. The two disciplines complement each other: you can keep one and add the other.

Go further: SEO agency or GEO specialist

Should I hire a freelance GEO consultant or an agency?

A freelancer fits a defined job: an audit, a strategy, a set of pages, with your team doing the rest. An agency fits ongoing work that needs measurement, writing, publishing and off-site distribution at the same time. Whichever you choose, ask for the same proof: a baseline on your questions, repeated runs, and a before/after measured with the same protocol.

Go further: What to ask for

How do I vet a GEO agency before hiring one?

Ask seven questions. Which buying questions will you measure, and who approves them? How many runs per question and engine? What noise threshold, measured how? Who writes, and who publishes, on my site and elsewhere? Do I get the raw answers? What commitment, and why? Can you show a before/after, with both measurements? Then check whether the agency itself shows up in AI answers.

Go further: The full vetting checklist

Can a GEO agency guarantee results in ChatGPT?

Not honestly. OpenAI itself says placement in ChatGPT search is not guaranteed [6], and answers vary from one run to the next [8]. An agency can commit to what it controls: the work produced and published, the measurement method, the raw data handed over, and objectives measured with the same protocol, reported even when they are missed.

Go further: Guarantees, in detail

Which AI search visibility agency works with cybersecurity companies?

Cybersecurity buyers (CISOs, IT leaders) ask precise technical and compliance questions: frameworks, managed services, alternatives to a vendor. A specialized agency knows those questions, the vocabulary and the sources AI engines reuse in this sector. ZivRank has a dedicated offer for B2B cybersecurity; its founder also founded Resilium, an affiliated cybersecurity company.

Go further: AI visibility agencies for cybersecurity

What is the best GEO agency for e-commerce brands?

For an online store, the questions that matter are product questions: best product for a use, one model versus another, a cheaper alternative. A good agency works your product pages (readable text, plain specifications, reviews), your feeds and structured data, and your presence on the comparisons and review sites that get cited. ZivRank has a dedicated retail and e-commerce offer.

Go further: GEO agencies for e-commerce brands

Is there a GEO agency that handles off-site distribution like review profiles, video and third-party mentions?

Few do the work themselves: most tracking tools recommend placements or let you book them, at your team’s initiative. Yet much of what gets cited lives outside your site. The ZivRank team produces and publishes third-party profiles, videos and satellite sites itself, funded with credits in its programs.

Go further: Off-site distribution, in detail

Which agency writes and publishes AI-optimized content for B2B brands?

GEO agencies that execute, as opposed to tools that provide briefs or drafts to review. Ask who writes, who checks the facts, and who publishes. At ZivRank, our experts write, fact-check and publish the content, then a re-measurement with the same protocol shows what moved.

Go further: Who writes and publishes

Are there white-label GEO services for marketing agencies?

Yes. Look for a partner that provides measurement, and production if needed, under your agency’s brand. Check three things: are reports in your brand, do prices leave you a margin, and who talks to the end client? ZivRank offers a white-label agency portal, with a partner price list shared on request and never published.

Go further: White-label GEO, in detail

Budget, timing and ROI

What GEO costs and what it returns.

How much does a GEO agency cost per month?

Published prices run from a few hundred to several thousand dollars a month, depending on how many questions are tracked and how much content is produced. One agency publishes an entry plan at €6,995 a month, for example. At ZivRank: free diagnostic, tracking from $390 per month, programs from $2,300 to $10,000 per month, over 6 months, 10% off when prepaid annually.

Go further: GEO pricing in 2026

What ROI can I expect from generative engine optimization, and how fast?

Visibility changes show after 3 to 8 weeks; the effect on pipeline takes longer. ROI is calculated on your numbers: leads from AI assistants, conversion rate and contract value, against the program cost. Without a baseline, no ROI can be shown. AI referral traffic is still small across the web, 0.32% of visits in SE Ranking’s 2026 sample [22], so count named answers, not only clicks.

Go further: GEO ROI and timing

Technical

What a site must allow to be read and reused by AI engines.

What should a technical GEO audit check?

Six things. AI crawler access in robots.txt and in CDN or firewall rules: OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot [5][14][15]. Text readable without JavaScript. No directive that blocks snippets. An answer block near the top and question-style subheadings. Valid structured data, a named author and an update date. An up-to-date sitemap submitted to Google and Bing, with IndexNow [18].

Go further: The ZivRank technical audit

How do I check if GPTBot, ClaudeBot and PerplexityBot can crawl my site?

Read your robots.txt for rules naming them, then check your CDN or firewall: bot-blocking settings can stop them even when robots.txt allows them. Know which bot does what: GPTBot and ClaudeBot collect training data, while OAI-SearchBot, Claude-SearchBot and PerplexityBot power search answers [5][14][15]. For visibility, the search bots matter most. Your server logs show whether they actually visit.

Go further: The ZivRank technical audit

Does llms.txt actually work for AI visibility?

llms.txt is a 2024 proposal: a Markdown file at the root of a site that summarizes it and lists its useful pages for language models [13]. It is cheap to create. However, the official documentation of the major engines reviewed for this FAQ does not describe using it, and no measured effect has been published. Add it if it is easy, without expecting results.

Go further: What our audit checks

Sources

Every figure in this article comes from the sources below, checked on November 9, 2026. Bracketed numbers in the text refer to this list.

  1. [1] Aggarwal et al. (arXiv; KDD ’24), GEO: Generative Engine Optimization, v3, June 28, 2024.
  2. [2] G2 (Tim Sanders), In the Answer Economy, Don’t Win the Click — Win the Answer, April 15, 2026.
  3. [3] Pew Research Center, Google users are less likely to click on links when an AI summary appears in the results, July 22, 2025.
  4. [4] Google Search Central, AI features and your website, last updated December 10, 2025.
  5. [5] OpenAI, Overview of OpenAI Crawlers, checked November 9, 2026.
  6. [6] OpenAI Help Center, Searching the web with ChatGPT, checked October 11, 2026.
  7. [7] OpenAI Help Center, Publishers and Developers FAQ, checked October 11, 2026.
  8. [8] SparkToro (Rand Fishkin), NEW Research: AIs are highly inconsistent when recommending brands or products, January 27, 2026.
  9. [9] Google Analytics Help, What’s new in Google Analytics (entry “New AI Assistant traffic measurement”), May 13, 2026.
  10. [10] Google Analytics Help, [GA4] Default channel group, checked October 11, 2026.
  11. [11] Google Search Central, Google Search’s guidance about AI-generated content, February 8, 2023.
  12. [12] Google Search Central, Spam policies for Google web search, checked November 9, 2026.
  13. [13] Jeremy Howard (Answer.AI), The /llms.txt file, 2024 proposal.
  14. [14] Perplexity, Perplexity Crawlers, checked November 9, 2026.
  15. [15] Anthropic, Does Anthropic crawl data from the web, and how can site owners block the crawler?, checked November 9, 2026.
  16. [16] Reddit, Reddit Rules, checked November 9, 2026.
  17. [17] Federal Trade Commission, Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, posted August 21, 2024.
  18. [18] IndexNow, IndexNow protocol, checked November 9, 2026.
  19. [19] Ahrefs (Despina Gavoyannis), Are AI Mode and AI Overviews Just Different Versions of the Same Answer?, December 15, 2025.
  20. [20] DerivateX, B2B SaaS AI citation study, May 31, 2026, updated September 22, 2026.
  21. [21] Ahrefs Help Center, AI Visibility Metrics, September 3, 2026.
  22. [22] SE Ranking (Yulia Deda), AI traffic grew 16x from 2024 to 2026, June 18, 2026.