# AEO & SEO Audits - Beanstalk Governance Studio Source: https://getbgs.com/aeo-seo {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":\[{"@type":"ListItem","position":1,"name":"Home","item":"https://getbgs.com"},{"@type":"ListItem","position":2,"name":"AEO & SEO","item":"https://getbgs.com/aeo-seo"}\]} AEO & SEO - AI Search Readiness # Get Cited by AI. Not Just Ranked by Google. Check real AI crawler access, score citation likelihood against research-backed criteria, and audit traditional SEO - all in one run, scored locally with an embedded language model that extends every check to 100 languages. **8**AI crawlers checked **7**dedicated reports **100**languages supported locally **0**page content sent to the cloud How It Works ## A multi-phase scan, not a single SEO checklist. Most SEO tools check titles and meta tags. Governance Studio runs a full pipeline covering both traditional SEO and the AI-crawler signals legacy tools were never built to measure. ### Six stages, one audit run Each stage builds on the last, from discovery through site-wide trust signals. STAGE 1 #### URL discovery Sitemap (with index-file and compressed-sitemap support), full crawl, or a manual URL list, with an automatic headless-browser fallback if a fast crawl gets blocked. STAGE 2 #### AI crawler access check robots.txt and llms.txt parsed for all 8 tracked AI bots, plus a multi-variant redirect consistency check (http/https, www/non-www). STAGE 3 #### Per-page content scan Concurrent headless Chromium page loads extract DOM structure, structured data, and real Core Web Vitals (TTFB/FCP/LCP/CLS), multi-run averaged to reduce noise. STAGE 4 #### Site health checks Broken links, broken images, missing alt text, and dead sitemap URLs. STAGE 5 #### On-page SEO & content structure Titles, meta descriptions, headings, thin-content detection, and how well content is structured for AI discoverability. STAGE 6 #### Site-wide E-E-A-T Experience, Expertise, Authoritativeness, and Trustworthiness signals, evaluated across the whole site. Every stage of this pipeline also works on authenticated pages - the same HTTP Basic Auth and raw header (cookie/authorization token) support used by the WCAG audit lets you check AI search readiness on gated content, staging, and localhost, not just public pages. AI Crawler Access ## 8 AI crawlers checked - not a guess. Real robots.txt and llms.txt parsing for every major AI crawler, with the retry and fallback handling needed to tell "blocked" apart from "couldn't verify." OpenAI GPTBot Powers ChatGPT's web browsing and training. Anthropic ClaudeBot Powers Claude's web access and training. Perplexity AI PerplexityBot Powers Perplexity's real-time answer citations. Gemini / Bard Google-Extended Controls use in Google's AI model training. Alexa / Rufus Amazonbot Powers Amazon's AI assistants and shopping tools. AI training sets CCBot Common Crawl - feeds many third-party AI models. Anthropic (alt.) anthropic-ai Anthropic's alternate crawler user-agent. Cohere AI cohere-ai Powers Cohere's AI models and products. Citation Scoring ## Research-backed, not a black box. Citation likelihood is scored with a real formula grounded in published research, not a guess - and it's honest about which signals matter most and which are still contested. ### What actually moves the score The two highest-weighted factors - quantitative data density and citing or quoting external sources - are drawn directly from Aggarwal et al., "GEO: Generative Engine Optimization," KDD 2024 (Princeton, Georgia Tech, Allen Institute, IIT Delhi) - described in the research as the only peer-reviewed study measuring this directly. A second factor, field-populated FAQ/Q&A schema, draws on Fischman (SSRN 2026), a 730-citation cross-platform study across ChatGPT and Gemini - with the scoring code explicitly noting this specific area is still contested in the research (a separate large-scale Ahrefs study found no effect at all). 2 Peer-reviewed/empirical studies cited directly in the scoring formula 100 Languages the scoring extends to via the embedded model 0 API calls required - citation scoring runs fully offline Embedded AI Model ## A real language model, built in - not bolted on. Governance Studio ships with its own multilingual language model, embedded directly in the application. No API key to configure, no account to create, no usage limits to track - it runs the moment you start an AEO/SEO audit. ### What it actually does The citation-scoring formula covered above is rule-based and research-cited - it does not depend on the embedded model to run. The model's real job is narrower and more useful: it extends those same detectors (question-format headings, author signals, attribution language) to work correctly on non-English and non-Latin-script sites, where plain-English keyword matching would otherwise detect nothing at all. 100 Languages supported, including non-Latin scripts ~118MB Model size, bundled into the application install 0 API calls or page content ever transmitted to run it **Fully offline, by design.** The model runs entirely on your desktop with no network dependency for scoring - remote model loading is disabled outright in the application, so a bundling issue fails closed instead of silently reaching out to a cloud API. Report Output ## 7 dedicated reports. One audit run. Not a single page bolted onto a WCAG audit - a real, separate documentation package your marketing or growth team can act on directly. .html Executive Summary Leads the package - overall AEO/SEO score and the highest-impact fixes across every dimension below. .html AI Crawler & Technical Report AI crawler access, robots.txt/llms.txt, schema, and sitemap health. .html AI Citation & Discoverability Report How likely each page is to be cited by AI engines like ChatGPT, Claude, or Perplexity. .html On-Page SEO Report Titles, meta descriptions, headings, thin content, and Core Web Vitals. .html Site Reliability Report Broken links, broken images, mobile-friendliness, mixed content, and HTTPS. .html Content Structure Report How well your content is structured for AI discoverability. .html E-E-A-T & Trust Report Experience, Expertise, Authoritativeness, Trustworthiness, and site-level trust signals. Pricing ## One license. Every AI search readiness audit. A single flat price covers unlimited AEO/SEO audits across every property you own - and includes WCAG and Site Health tools in the same license, not a separate add-on. Governance Studio License $399 / month $4,450 billed annually - or month-to-month at $399 - Unlimited properties - no domain or site cap - Unlimited pages - no volume tiers or overages - Unlimited audit runs - scan as often as needed - 5 simultaneous sign-ins, each with their own login - All 7 AEO/SEO report formats included - 8 AI crawlers checked - GPTBot, ClaudeBot, PerplexityBot, and 5 others - Research-backed citation-likelihood scoring - Embedded multilingual language model - 100 languages, fully offline - Also included: 11 WCAG report formats, jurisdiction-correct VPAT® - Also included: Site Health tools - broken links, regex, meta, images - Localhost, staging & authenticated-page scanning - 7-day free trial - no credit card required Common Questions ## Things teams ask about AEO & SEO. Does the embedded AI model score my content, or is it marketing? + Citation scoring itself is a transparent, rule-based formula grounded in published research - not an AI judgment call. The embedded model's real job is narrower and more useful than "AI scores your content": it extends the same rule-based detectors to work correctly on non-English and non-Latin-script sites, where plain-English keyword matching would otherwise detect nothing. It runs fully offline, with no page content ever transmitted anywhere. Is 8 crawlers enough, or are you missing major AI bots? + The 8 tracked crawlers cover OpenAI (GPTBot), Anthropic (ClaudeBot and the alternate anthropic-ai user-agent), Perplexity, Google's Gemini/Bard training crawler (Google-Extended), Amazon (Amazonbot), Common Crawl (CCBot, which feeds many third-party models), and Cohere. This list reflects the current major AI crawler landscape and gets reviewed as new bots become relevant. Do you check real Core Web Vitals, or an estimate? + Real, in-page measurement via the browser's own Performance Observer API - TTFB, FCP, LCP, and CLS - captured during an actual headless-browser page load, with multiple runs averaged to reduce dev-machine timing noise rather than a single, potentially misleading sample. Can this replace our existing SEO platform? + It's built to run alongside your existing SEO stack, not necessarily replace it - though for a team on a per-domain-metered enterprise SEO platform, the flat $399/month price alone often beats the cost of either tool on its own. What it specifically adds is the AEO layer most SEO platforms don't have at all: real AI crawler access checking and research-backed citation-likelihood scoring. {"@context":"https://schema.org","@type":"FAQPage","mainEntity":\[{"@type":"Question","name":"Does the embedded AI model score my content, or is it marketing?","acceptedAnswer":{"@type":"Answer","text":"Citation scoring itself is a transparent, rule-based formula grounded in published research - not an AI judgment call. The embedded model's real job is narrower and more useful than \\"AI scores your content\\": it extends the same rule-based detectors to work correctly on non-English and non-Latin-script sites, where plain-English keyword matching would otherwise detect nothing. It runs fully offline, with no page content ever transmitted anywhere."}},{"@type":"Question","name":"Is 8 crawlers enough, or are you missing major AI bots?","acceptedAnswer":{"@type":"Answer","text":"The 8 tracked crawlers cover OpenAI (GPTBot), Anthropic (ClaudeBot and the alternate anthropic-ai user-agent), Perplexity, Google's Gemini/Bard training crawler (Google-Extended), Amazon (Amazonbot), Common Crawl (CCBot, which feeds many third-party models), and Cohere. This list reflects the current major AI crawler landscape and gets reviewed as new bots become relevant."}},{"@type":"Question","name":"Do you check real Core Web Vitals, or an estimate?","acceptedAnswer":{"@type":"Answer","text":"Real, in-page measurement via the browser's own Performance Observer API - TTFB, FCP, LCP, and CLS - captured during an actual headless-browser page load, with multiple runs averaged to reduce dev-machine timing noise rather than a single, potentially misleading sample."}},{"@type":"Question","name":"Can this replace our existing SEO platform?","acceptedAnswer":{"@type":"Answer","text":"It's built to run alongside your existing SEO stack, not necessarily replace it - though for a team on a per-domain-metered enterprise SEO platform, the flat $399/month price alone often beats the cost of either tool on its own. What it specifically adds is the AEO layer most SEO platforms don't have at all: real AI crawler access checking and research-backed citation-likelihood scoring."}}\]} ## See your own AI crawler access score. Download the app, create your free trial account, and run a real AEO/SEO audit on one of your own properties. No call required. [Download Free Trial](/#cta) $399/mo - Unlimited properties - 7 AEO/SEO reports - 7-day free trial - no credit card required