How to Get Mentioned in ChatGPT, Claude & Gemini Answers — 20 Proven Steps

1Why AI Mentions Matter More Than Rankings

When someone asks ChatGPT "best project management tool for startups," it doesn't show 10 blue links. It gives one curated answer: 3-5 products with descriptions, pricing, and pros/cons. If your product isn't in that answer, you don't exist for that buyer.

This is fundamentally different from SEO. Ranking #7 on Google still gets you clicks. Being absent from an AI answer gets you zero. And research from Search Engine Land shows that 50% of AI-cited sources come from pages ranking in positions 4-10 — you don't even need to rank #1 to get cited, but you do need the right signals.

The good news: the signals that make AI recommend you are specific, measurable, and fixable. Every step in this guide is something you can implement — most in under an hour.

2Run a Baseline AI Visibility Audit

Before fixing anything, measure your starting point. An AI visibility audit scans your site across 25+ signals in 6 categories and returns a score from 0-100:

• AI Crawl Access — Can GPTBot, ClaudeBot, PerplexityBot, and Google-Extended reach your pages?
• Content Quality — Is your content structured for AI extraction? Facts-first writing?
• Product Clarity — Does AI know what you do, who you serve, and what you cost?
• Structured Data & Meta — JSON-LD, Schema.org, Open Graph tags?
• Agent Readiness — Do you have llms.txt, llm.json, FAQ schema, sitemap?
• Trust & Social Proof — Do third parties mention and validate you?

You also get your top blockers ranked by impact, a quick wins list, and auto-generated fix files (llms.txt, JSON-LD) ready to deploy.

3Open Your Site to AI Crawlers (robots.txt)

Over 40% of websites accidentally block AI crawlers. Your robots.txt might be preventing ChatGPT, Claude, Perplexity, or Gemini from reading your content. If AI can't crawl your site, it relies on outdated training data — which often means it can't describe or recommend you at all.

Common blocking mistakes: Cloudflare's default bot protection can block AI crawlers. Vercel and other platforms sometimes inherit restrictive rules. Password-protected staging content, JavaScript-heavy SPAs, and paywalls also prevent crawling.

Add these lines to your robots.txt:

# Allow AI crawlers
User-agent: GPTBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: OAI-SearchBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: meta-externalagent
Allow: /

GPTBot and ChatGPT-User are OpenAI's training and real-time search crawlers. OAI-SearchBot is their newest search-specific bot. ClaudeBot and Claude-SearchBot are Anthropic's. PerplexityBot powers Perplexity's live citations. Google-Extended feeds Gemini. meta-externalagent is Meta AI.

4Create llms.txt — Your AI Cheat Sheet

llms.txt is a plain text file at your website root (yourdomain.com/llms.txt) that tells AI crawlers exactly what your product does. Think of it as robots.txt for AI — but instead of saying what to block, it says what to know.

Without it, AI systems guess from your HTML and often get it wrong. With it, they cite you accurately. This is the single most impactful file for AI visibility — especially for Claude, which is one of the most llms.txt-aware models.

# Your Product Name

> One-sentence description of what your product does.

## Key Facts
- Website: yourdomain.com
- Category: SaaS / Your Category
- Pricing: Free plan, Starter $29/mo, Pro $79/mo
- Founded: 2025
- Target audience: Who this is for

## What We Do
2-3 sentences describing the core value proposition.
Focus on facts, not marketing language.

## Key Features
- Feature 1 — short description with specific benefit
- Feature 2 — short description with specific benefit
- Feature 3 — short description with specific benefit

## How It Works
1. Step one of the core workflow
2. Step two
3. Step three

## Integrations
- Integration 1
- Integration 2

## Resources
- Pricing: https://yourdomain.com/pricing
- Documentation: https://yourdomain.com/docs
- Blog: https://yourdomain.com/blog
- Changelog: https://yourdomain.com/changelog

Also create llms-full.txt with extended detail — full feature descriptions, API documentation, integration guides, and use cases. AI systems will read whichever version they find.

5Add JSON-LD Structured Data

JSON-LD tells AI systems your product type, pricing, features, and ratings in a machine-readable format. Sites with proper structured data are significantly more likely to be cited accurately in AI answers. According to Princeton's GEO research, structured citations can increase AI visibility by up to 40%.

Add these three Schema.org types to your homepage:

• SoftwareApplication — name, description, applicationCategory, offers (with real prices), featureList, aggregateRating
• Organization — name, URL, logo, sameAs (social links), foundingDate, contactPoint
• FAQPage — common questions about your product with direct answers

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "SoftwareApplication",
  "name": "Your Product",
  "url": "https://yourdomain.com",
  "applicationCategory": "BusinessApplication",
  "operatingSystem": "Web",
  "description": "Specific one-line description with category + audience",
  "offers": [
    { "@type": "Offer", "price": "0", "priceCurrency": "USD", "name": "Free" },
    { "@type": "Offer", "price": "29", "priceCurrency": "USD", "name": "Starter" },
    { "@type": "Offer", "price": "79", "priceCurrency": "USD", "name": "Pro" }
  ],
  "featureList": ["Feature 1", "Feature 2", "Feature 3"],
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.8",
    "reviewCount": "127"
  }
}
</script>

6Add FAQ Schema Markup

FAQPage schema is the #1 most-cited structured data type by AI assistants. When you add FAQ schema, AI systems can directly use your Q&A pairs as answers — they literally copy your text.

Write 7-10 questions covering:

• "What is [your product]?" — The definitional question AI gets asked most
• "How much does [your product] cost?" — Pricing is the #1 follow-up question
• "How is [your product] different from [competitor]?" — Comparison queries drive buying decisions
• "Who is [your product] best for?" — Audience targeting helps AI match recommendations
• "Does [your product] have a free plan?" — Free tier is a key decision factor
• "What integrations does [your product] support?" — Integration compatibility matters
• "How do I get started with [your product]?" — Onboarding simplicity

These must be added as FAQPage JSON-LD in your page's <head>, not just displayed as visible text. AI specifically looks for this schema type. According to Moz research, content with clear answer boxes gets featured up to 40% more often in AI-generated responses.

7Write Answer-First Content

AI doesn't read marketing copy — it extracts facts. Research by Kevin Indig analyzing 7,000 AI citations found that content depth and readability have the biggest impact on whether AI cites a source. Vague claims get skipped. Specific facts get cited.

The Wikipedia rule: Write like Wikipedia — lead every section with a direct definition or fact. Wikipedia provides 47.9% of ChatGPT's citations for a reason: it leads with the answer, not the buildup.

Rules for AI-friendly content:

• Lead every section with a specific fact or answer in the first sentence
• Use HTML tables for pricing and feature comparisons — AI extracts tables 30% more reliably than prose
• Use H2/H3 headings that match what people ask AI: "How much does X cost?" "What is X best for?"
• Include specific numbers everywhere: user counts, pricing, performance metrics, percentages
• Add a "Quick Answer Box" at the top of key pages: product name, best for, cost, key benefit

8Build a Clear, Specific Homepage

Your homepage H1 is the first thing AI crawlers read. "Welcome to our platform" tells AI nothing. "Project management for remote engineering teams — track sprints, ship faster" tells AI exactly what you are, who you serve, and what you do.

Three homepage essentials for AI:

• Specific H1 — Include your category, audience, and value prop. This is what AI uses for entity recognition.
• Crawlable pricing page — AI is constantly asked "how much does X cost?" If your pricing is behind a login, in a JavaScript modal, or missing, AI can't answer. Create a /pricing page with static HTML showing plan names, prices, and features in a table.
• Product description in the first 200 words — AI reads top-of-page content first. Don't bury your product description below the fold or behind animations.

9Create Comparison & Alternative Pages

When users ask AI "X vs Y" or "best [category] tool," it looks for comparison pages first. A single well-structured comparison page with a feature table can drive more AI citations than your entire blog combined.

Create pages like:

• /compare/yourproduct-vs-competitor — side-by-side feature table, pricing comparison, and "best for" section
• /alternatives — list alternatives in your category including yourself, with honest pros/cons for each
• /compare/category-tools — broader category roundup

Be honest. Include cases where competitors might be better for certain use cases. AI trusts balanced comparisons — one-sided marketing pages get deprioritized. If your comparison page says "we're best at everything," AI knows that's not credible.

10Build Your SEO Foundation

SEO and AI visibility are complementary, not competing. Search Engine Land's research shows that 50% of AI-cited sources also rank in Google's top 10. Strong Google rankings feed into ChatGPT citations (via Bing) and Gemini citations (via Google's index).

The essential SEO checklist for AI visibility:

• Keyword research — Target queries with 100+ monthly searches. Mix commercial ("best [category] tool"), informational ("how to [solve problem]"), and long-tail ("[product] vs [competitor] for [audience]") keywords.
• On-page optimization — Title tags with primary keyword, intent-matching H2/H3 headers, internal linking between related pages, canonical URLs.
• Technical health — Set up Google Search Console AND Bing Webmaster Tools. Submit your XML sitemap to both. ChatGPT uses Bing for web search, so Bing indexing directly impacts ChatGPT visibility.
• Backlink building — Guest posts on industry blogs, directory listings (Product Hunt, G2, Capterra), publishable research that earns links.

11Dominate Reddit

Reddit is one of the highest-ROI channels for AI visibility. OpenAI has a direct data partnership with Reddit. Reddit threads rank in Google's top 3 results for thousands of queries. SparkToro's analysis shows Reddit posts can rank in Google within 24 hours and maintain visibility for years.

Why Reddit matters for AI:

• 430M+ monthly active users generating real discussions AI trains on
• OpenAI-Reddit partnership ensures Reddit content feeds directly into ChatGPT
• Google indexes Reddit heavily, feeding into Gemini citations
• Perplexity frequently cites Reddit threads as sources

Target subreddits by audience size:

Reddit posting strategy:

• Build credibility first. Comment helpfully on 10-15 threads before posting. Earn karma. Read subreddit rules carefully — many ban promotional posts.
• Posts generate 10x more visibility than comments. Write analysis posts ("I analyzed 100 SaaS landing pages — here's what works"), step-by-step guides, or honest case studies.
• Mention your product only when directly relevant — roughly 10% of your activity. "I built [product] to solve this" is fine if the post genuinely helps. "Check out [product]!" gets downvoted and removed.
• Use data-driven edits. After posting, add an edit with updated data or results. This keeps the post active and signals authenticity.

12Create YouTube Content

YouTube is the most underrated AI visibility channel. Google owns YouTube and gives preferential treatment to YouTube transcripts in Gemini's answers and Google AI Overviews. HubSpot data shows 87% of marketers report video increases website traffic.

Why YouTube wins for AI visibility:

• Google ownership means YouTube transcripts feed directly into Gemini
• YouTube transcripts provide text for AI to extract and cite
• YouTube videos get shared 12x more than text content
• Less competition than blog content in most categories

Video formats that get AI citations:

• "How to [solve problem] in 2026" — tutorial format
• "[Your tool] vs [Competitor] — honest comparison"
• "[Industry] tools ranked — which is best for [audience]"
• "Complete guide to [topic]" — comprehensive walkthrough
• "5-minute demo of [Your Product]" — product showcase

YouTube optimization checklist for AI:

• Title: Include year + primary keyword + benefit. AI uses titles for query matching.
• Timestamps: Add chapter timestamps to your description. AI uses these to match specific queries to specific sections of your video.
• Description: Write 500+ words minimum. First 125 characters are visible in search results — front-load your keyword. Include your product URL, feature list, and pricing.
• Tags: Add 10-15 relevant tags. Main keyword first, then competitor names, "vs" and "alternative" variations.
• Transcript: Clean up the auto-generated transcript. Add punctuation, fix product names, and ensure it reads well as text — because AI reads transcripts as text.

You don't need a studio. Screen recording with Loom or OBS (free), editing with DaVinci Resolve (free), thumbnails in Canva (free), and noise removal with Krisp.ai. A laptop mic is fine for demos.

13Build Third-Party Authority

AI models check external sources before recommending. If nobody else mentions you, the AI won't either. Particularly for ChatGPT (which searches the web) and Gemini (which uses Google's index), third-party mentions are a strong trust signal.

Focus on quality over quantity. One detailed Product Hunt launch with 100+ upvotes and real customer comments is worth more than 50 thin directory listings.

14Get Added to Content AI Already Cites

This is one of the highest-leverage tactics most people miss. When ChatGPT or Perplexity answers a query about your category, it cites specific sources — blog posts, comparison articles, review roundups. If you can get your product mentioned in those already-cited sources, you inherit their AI visibility immediately.

The process:

1. Ask ChatGPT and Perplexity your key buyer prompts ("best [category] tools for [audience]")
2. Document every URL the AI cites in its response
3. Check each cited article — is your product mentioned? If not, that's your target.
4. Find the author via the article byline, LinkedIn, or Twitter
5. Reach out and ask to be included — offer a product demo, unique data, or a guest quote

Focus on "money prompts" — specific, buyer-intent queries like "best email marketing tool for ecommerce under $50/mo," not broad terms like "email marketing." The more specific the prompt, the higher the buying intent and the easier it is to get included in a niche article.

15Launch Strategic PR Campaigns (HARO)

Press mentions create a powerful chain: media coverage gets indexed by Google, AI crawlers read indexed articles, and your brand gets associated with authority in your category. TechCrunch, Forbes, or even industry-specific publications can dramatically accelerate AI visibility.

The easiest PR channel: HARO (Help a Reporter Out)

1. Sign up free at HARO (now Connectively)
2. You receive 3 daily emails with journalist queries looking for expert sources
3. Respond to 3-5 relevant queries per week with a 2-3 paragraph expert answer
4. Include your credentials and an offer for follow-up quotes

HARO's own data shows a 5-10% publication success rate per response. That means 5 weekly responses could land you 1-2 press mentions per month. Over 6 months, that's 6-12 authoritative publications mentioning your brand — more than enough to move the needle in AI recommendations.

Content that attracts AI attention via PR:

• Original research or survey data ("We analyzed 1,000 SaaS websites and found...")
• Expert commentary on industry trends
• Problem-solving case studies with specific results
• Thought leadership on emerging topics (AI, automation, new regulations)
• Contrarian takes backed by data

16The Wikipedia Path

Wikipedia is one of the most heavily cited sources by AI. But creating a Wikipedia page requires meeting notability guidelines — you need 5+ independent reliable sources covering your company (not press releases, not your own blog).

The path to Wikipedia eligibility:

1. Earn 5+ press mentions in notable publications (TechCrunch, Forbes, industry journals) via HARO and PR outreach
2. Ensure coverage goes beyond press releases — independent journalist analysis is required
3. Once you have sufficient coverage, create a neutral, well-sourced Wikipedia draft
4. Follow Wikipedia's conflict of interest guidelines — disclose your affiliation

This is a long-term play (6-12 months), but the payoff is enormous. A well-maintained Wikipedia page becomes a permanent, high-authority citation source that every AI model trusts.

17Add Social Proof AI Can Parse

AI assistants factor in trust signals when deciding what to recommend. But the proof needs to be in the HTML — not in images, JavaScript-rendered carousels, or PDFs that crawlers can't read.

What works:

• Testimonials with real names and companies in plain HTML text
• User counts in a visible heading: "Trusted by 5,000+ teams"
• Customer logos as <img> tags with descriptive alt text (e.g. "Stripe logo — customer")
• Review ratings linked to aggregateRating in your JSON-LD
• Case study links with specific results in the anchor text

What doesn't work:

• Logos as a single background image with no alt text
• Testimonials rendered by JavaScript after page load
• Social proof only visible in downloadable PDFs or slide decks
• "Trusted by leading companies" without naming them

18Run a Security Check

AI systems also evaluate whether your site is trustworthy. Exposed API keys, missing security headers (CSP, HSTS, X-Frame-Options), or leaked .env files signal that your product isn't production-ready — and AI deprioritizes untrustworthy sites.

19Platform-Specific Tactics

The foundation is the same across all platforms. But each AI has unique signals it prioritizes.

ChatGPT tips

• Submit your sitemap to Bing Webmaster Tools — ChatGPT's web search uses Bing
• Ensure your structured data is rich — ChatGPT extracts JSON-LD heavily
• Reddit posts get surfaced via the OpenAI-Reddit partnership

Claude tips

• Make your llms.txt as detailed as possible — Claude is the most llms.txt-aware model
• Write clear, factual prose — Claude deprioritizes marketing language
• Entity clarity in your H1 and first paragraph matters enormously

Gemini tips

• Create YouTube content — Gemini weighs YouTube transcripts more than any other AI
• Update your Google Business Profile if you have one
• Strong Google SEO directly feeds into Gemini citations via Google's search index

Perplexity tips

• Perplexity uses live web search with citations — freshness matters
• Get mentioned on pages Perplexity already cites for your key queries
• Perplexity frequently cites Reddit threads and YouTube videos

20Track & Monitor Your AI Visibility Over Time

A one-time scan tells you where you stand. But AI answers change constantly — competitors gain visibility, models update, and your fixes need validation. You need continuous monitoring across all platforms.

What to monitor:

• Which of the 7 AI platforms (Perplexity, ChatGPT, Gemini, Google AI Overviews, Google AI Mode, Microsoft Copilot, DeepSeek) mention you — and which don't
• Which competitors appear instead of you for your key buyer prompts
• How your visibility score changes week over week
• What content gaps to fill based on prompts where competitors appear but you don't
• Sentiment — is AI describing your product positively, neutrally, or negatively?

Timeline: What to Expect

AI visibility compounds over time. Here's a realistic month-by-month timeline:

The Complete Checklist

Every action from this guide condensed into one list. Work through it top to bottom.

Frequently Asked Questions