Which AI engines does SeenRank track?

ChatGPT, Claude, Perplexity, and Google AI Overview. We run real queries against each one weekly and parse the responses for whether your business is mentioned, where, and what was said about you.

How is this different from regular SEO tracking?

Regular SEO tools check Google's ten blue links. SeenRank checks the answers AI engines now give above and beside those links. Different surface, different ranking signals, different game.

How are queries chosen?

When you add a business, SeenRank suggests 20 queries based on your category and audience, the kind of buying-intent questions real customers type into ChatGPT or Claude. You can edit the list, add your own, or use ours.

What does the why-you-weren't-cited report do?

Each week, Claude reads the AI responses your business missed and writes three specific, concrete recommendations. Things like publish a comparison page on this exact topic or answer this question on your FAQ.

Can I run this on multiple businesses for clients?

Yes. Agency plans add white-label PDF reports, a shareable client dashboard, bulk CSV import, and a weekly digest summarizing every client. The Agency+ plan adds API access and a custom subdomain.

Do you store the raw AI responses?

Yes. We store the full raw response for every query so you can audit exactly what the AI said and screenshot it for clients. Stored securely in Supabase.

Does this work for SaaS, B2B, e-commerce, or service businesses?

Yes. SeenRank works for any business in any category. Local services, SaaS, e-commerce, B2B, agencies, creators. If a customer might ask an AI engine about your category, we track whether you're cited.

One click in the billing page. No phone calls, no please-don't-leave emails. You can also downgrade or pause.

How long does the full GEO checklist take to work through?

Per-page items on 5 pages: roughly one afternoon. Sitewide infrastructure: 2-3 hours one-time. Off-site brand work: ongoing, 2-3 hours per week. Measurement: 15 minutes per week once set up. Total first-week investment: 8-10 hours.

SeenRank Blog

The 2026 GEO checklist

Updated 2026-05-13. By the SeenRank team.

Short answer: the 27 checks below are the tactical distillation of what wins citations inside ChatGPT, Claude, Perplexity, and Google AI Overview in 2026. Items are grouped by leverage: per-page (highest), sitewide infrastructure, off-site brand work, and measurement. Run through them in order. Most operators are missing items 1-3, which is also where 80% of the visibility lift comes from.

How to use this checklist

Don't try to fix everything at once. Three phases work for almost every team:

Week 1: Run a free check to baseline. Audit the per-page section (items 1-12) against your top 5 highest-revenue pages. Fix the gaps you find.
Weeks 2-3: Sitewide infrastructure (items 13-18). One-time work that compounds across every page on the site.
Ongoing: Off-site brand work (items 19-22) and measurement (items 23-27). Brand work is the slow-compounding multiplier; measurement is how you know any of this is working.

For context on each item, the long-form explanations live in the 2026 AI Search Visibility guide and the related posts linked inline.

Per-page checks (highest leverage)

These run against each individual content page. Audit your top 5 pages first, then the next 10. Most "I'm not getting cited" problems trace to items 1-3.

1. The first 200 words directly answer the buying-intent question

40-55 word direct answer in the opening paragraph, no preamble, no "in this article we'll cover". AI engine extractors evaluate the opening heavily. This is the single most common gap on otherwise-strong pages.

2. At least 2-3 statistics with cited authoritative sources

The Princeton KDD 2024 study measured +41% citation lift from adding statistics. Sprinkle them across the body, not bunched at the end. Each stat is a potential pull-quote.

3. At least one direct quote from a named expert

+28% citation lift in the Princeton study. Yourself counts if you have credentials in the topic. Use full attribution.

4. Question-based H2 and H3 headers

"How does X work?" beats "About X" or "X overview". AI engines map the user prompt to your headers. Match the way real customers phrase the question.

5. FAQ section with FAQPage JSON-LD

Highest yield, lowest effort. Any Q&A section gets FAQPage schema. AI engines extract these directly as candidate answers.

6. Article JSON-LD with author, datePublished, dateModified, mainEntityOfPage

Every long-form content page should have Article schema. Include a linked Person as author with their LinkedIn in sameAs for E-E-A-T signal.

7. HowTo JSON-LD on step-by-step content

For procedural pages. Voice assistants and Perplexity both lean on this. Use real step objects, not just a numbered list in prose.

8. `dateModified` reflects real recent updates

Perplexity especially cites content updated in the last 3 months at roughly 2x the rate of older content. Update dateModified when content genuinely changes. Don't fake updates.

9. Original element on the page

AI engines deduplicate aggressively. Each page needs something unique: original data, original screenshots, your own opinion, a chart you built. Restating Wikipedia in your own words gets you nowhere.

10. Brand name appears naturally throughout

Don't hide your own brand. AI engines learn association from co-occurrence on your own page as well as the wild. Mention yourself by name (not just "we") in headlines and key paragraphs.

11. Internal links to 2-3 related cluster pages

With descriptive anchor text. Hub-and-spoke internal-link structure helps AI engines map your topical authority. Avoid generic "click here" links.

12. Core Web Vitals pass mobile

LCP under 2.5s, INP under 200ms, CLS under 0.1. Slow pages get partial fetches from AI crawlers. Run PageSpeed Insights on every page on the audit list.

Sitewide infrastructure (one-time, compounds)

Set once, benefit on every page. Most of these are 30 minutes of work that you'll never have to touch again.

13. robots.txt allows GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot, Applebot-Extended

Open /robots.txt in your browser. Confirm all six bots have Allow: /. Blocking Google-Extended does not block regular Google indexing, only AI Overview and training pathways. Most sites should allow.

14. llms.txt manifest at /llms.txt

Markdown file pointing AI agents at your most important content. Format: a one-paragraph site summary, then sections of bulleted links to your key pages. Adoption is still early but cost is near zero.

15. XML sitemap up to date and submitted to Google Search Console

Helps Google's index speed, which feeds AI Overview. Submit a fresh sitemap after major site updates.

16. Organization JSON-LD with sameAs links

Site-wide Organization schema with sameAs array linking to your LinkedIn, X, GitHub, and any Wikipedia entry. Helps AI engines connect your site to your brand entity.

17. WebSite JSON-LD with SearchAction

Enables the sitelinks search box in Google and signals to AI engines that your site is structured. Standard pattern, copy-paste from schema.org.

18. .well-known/openapi.json and .well-known/ai-plugin.json if you have an API

Lets AI agents discover programmatic access to your data. Not relevant if you don't have an API; high-leverage if you do.

Off-site brand work (slow-compounding multiplier)

This is the GEO-specific work that SEO doesn't require. AI engines weight third-party brand mentions heavily, far more than they weight backlinks. The ROI horizon is months, not weeks.

19. Genuine Reddit answers in 2-3 niche subreddits per week

Where your audience actually hangs out. Be useful first, mention your brand only when directly relevant. One useful answer per week per subreddit beats ten promotional ones.

20. Quora answers on the top 10 questions in your category

Long-form, evergreen. Quora gets crawled by every AI engine. Single answers compound for years.

21. LinkedIn long-form posts 2-3 times per week from a real person

From your founder, marketing lead, or domain expert. AI engines crawl LinkedIn heavily for B2B brand signal. The voice matters: first-person, opinion-bearing, no engagement-bait.

22. One podcast appearance per quarter

Pitch the show notes mention as a precondition. Podcast show notes are crawled by AI engines and provide both a brand mention and a linkable citation.

Measurement (how you know any of this is working)

SEO has Google Search Console. GEO needs its own measurement stack. Four metrics matter, tracked weekly.

23. Mention Rate weekly across 4 engines

What percentage of AI answers mention your brand name. Track across ChatGPT, Claude, Perplexity, and Google AI Overview for your top 20 buying-intent queries.

24. Citation Rate weekly across 4 engines

What percentage of AI answers include a clickable URL to your domain. Different from Mention Rate: you can be mentioned without being cited (and vice versa).

25. Position in the answer

Where in the synthesized answer your citation appears. Position #1 mentions correlate with roughly 5x more downstream branded searches than position #5.

26. Prompt Coverage

How many of your 15-25 target prompts you appear in at all. A leading indicator of category-level visibility.

27. AI traffic attribution (referrer logs + pixel)

Filter analytics by AI-engine hostnames (perplexity.ai, chat.openai.com, claude.ai, gemini.google.com, copilot.microsoft.com). For ChatGPT dark traffic, install an AI traffic pixel. See how to find which AI engines send you traffic.

The 5-minute version of this checklist

If you only have 5 minutes today, do these three things in this order:

Run a free check to baseline (/check). 30 seconds.
Open your single highest-revenue page. Read the first paragraph aloud. If it doesn't directly answer the buying-intent question that page targets in 40-55 words, rewrite that paragraph. 5 minutes.
Open /robots.txt on your site. Confirm GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot, and Applebot-Extended are all allowed. 30 seconds.

Those three moves cover roughly half the lift of the full 27-item checklist. The rest is execution over weeks.

Run a free SeenRank check now →

FAQ

How long does the full checklist take to work through?

Per-page items (1-12) on 5 pages: roughly one afternoon if you have the writing skill in-house. Sitewide infrastructure (13-18): 2-3 hours one-time. Off-site brand work (19-22): ongoing, 2-3 hours per week. Measurement (23-27): 15 minutes per week once set up. Total first-week investment: 8-10 hours. Ongoing: 3-4 hours per week.

Which 3 items have the largest measured lift?

Per the Princeton KDD 2024 study: statistics added to content (+41%), direct quotes (+28%), authoritative source citations (large but unquantified lift). All three are in the per-page section (items 2, 3, and embedded in item 9). Most operators are missing all three.

I don't have time to do all 27. Where do I start?

The 5-minute version in the section above. After that, items 1-3 on your top 3 pages. That's 80% of the lift in 2-4 hours of focused work.

Is there a one-tool way to track items 23-27?

Yes, that's the SeenRank product. SeenRank tracks Mention Rate, Citation Rate, Position, and Prompt Coverage across ChatGPT, Claude, Perplexity, and Google AI Overview every Monday for every business you add. Free check tool gives you one data point in 30 seconds.

Will this checklist still be accurate in 12 months?

Items 1-22 are likely durable; they're about content quality and infrastructure, which AI engines have rewarded consistently across multiple model generations. Items 23-27 (measurement) will evolve as new tools mature and as AI engines publish more transparent analytics. Recheck the GEO note in your knowledge base quarterly.