AI Search Visibility Audit

Measurement of where your brand appears across ChatGPT, Perplexity, Gemini, Copilot, and AI Overviews, plus the technical and off-site reasons you're missing from the answers you care about. Delivered with a prioritized fix list and referral tracking set up so you can measure what changes.

What's included

Crawler access

• robots.txt review for GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, Claude-SearchBot, Google-Extended, and Applebot-Extended
• CDN and WAF check, since bot rules on Cloudflare and similar block AI crawlers by default on plenty of setups
• Server log analysis showing which AI crawlers actually reach you and what they fetch
• Rendering check, because most AI crawlers don't run JavaScript and a client-rendered app hands them an empty page
• Bing indexation review, since Copilot runs on the Bing index

Visibility measurement

• Prompt set built across branded, unbranded, comparison, and problem-stage queries, with the selection method documented
• Repeat sampling across multiple runs and regions, with variance reported next to every score
• Model and date recorded per result
• Prompt-level breakdown of where you're mentioned, cited, or absent
• Competitor gap analysis showing the prompts and sources where rivals appear and you don't
• Sentiment and accuracy review, including cases where a platform states wrong pricing or describes products you discontinued

Sources and citations

• Citation audit of which domains these platforms pull from in your category
• Third-party surface review across Reddit, YouTube, G2, Capterra, Wikipedia, and the listicles ranking for your terms
• Overlap analysis between AI Overview citations and classic organic top ten results
• Brand entity consistency across Wikidata, knowledge panel, and directory listings

Content

• Gap analysis tied to the prompts your buyers use, mapped to pages that would answer them
• Comparison and alternatives page coverage, since those formats get pulled more than anything else
• Freshness, author attribution, and original data on the pages you want cited
• Schema and content structure review

Measurement setup

• Referral tracking configured in GA4 and PostHog for chatgpt.com, perplexity.ai, gemini.google.com, and copilot, with a custom channel group so it stops landing in unassigned
• Conversion reporting on AI referral traffic
• Baseline showing AI referrals as a share of total traffic, recorded before any work starts
• Prioritized action list ranked by expected impact

Why it matters

Most of what's sold as AI optimization is guesswork with a dashboard attached. Nobody outside these companies knows the ranking mechanics, prompt results shift between runs, and a visibility score with no variance next to it tells you nothing. So this audit spends most of its time on things that are verifiable. Whether crawlers can reach your pages. Whether your content renders without JavaScript. Which third-party sources get cited in your category. How much traffic and revenue AI referrals actually send you today.

The unglamorous finding is that AI Overviews lean on the same index and signals as regular search, and the citations in most categories go to sources you don't own. Which means the work is usually a mix of ordinary technical SEO, getting mentioned in the places already being cited, and publishing something worth quoting. That's less exciting than a new acronym, and it's what moves the number.