
The DIY AI Visibility Audit: 12 Checks to See Where Your Business Stands

Table of Contents
The DIY AI Visibility Audit: 12 Checks to See Where Your Business Stands #
If your business depends on organic search discovery, you are losing traffic to AI-generated answers right now. Over the past six months, I have audited dozens of client sites and found that up to 40% of informational search traffic has migrated directly into ChatGPT, Perplexity, and Google AI Overviews. If your brand is not being cited in those answers, you do not exist to those buyers.
I am William Spurlock — AI Solutions Architect, Fractional AI CTO, and SEO-certified since 2021. I build AI-native automation stacks and premium web experiences, and I can tell you that traditional SEO is no longer enough. You need to know exactly how answer engines see your business. This post is a hands-on, zero-budget DIY AI visibility audit you can run in a single afternoon to find out exactly where you stand.
Why Do You Need a DIY AI Visibility Audit Today? #
You need a DIY AI visibility audit because answer engines are actively replacing traditional search results, and if you do not measure where your brand is cited, you cannot optimize for discovery. Traditional rank tracking is blind to the AI-generated summaries that now capture the majority of user clicks.
If you have noticed a sudden dip in your organic click-through rates, you need to determine whether Google's new search interface is the culprit. I wrote a guide on how to tell if Google AI Overviews caused your traffic drop to help you isolate this exact issue. Once you know where the leak is, you can start tracking how to measure AI visibility and the metrics that actually matter in 2026 to establish a baseline.
Answer engine optimization (AEO) is not about keyword density or backlinks. It is about structuring your brand facts and content so that LLMs can extract, verify, and cite them. This audit gives you a clear diagnostic map of your business's AI footprint without spending a dollar on expensive enterprise tools.
How Do You Audit Direct Citations in ChatGPT, Perplexity, and Gemini? #
You audit direct citations by querying the major LLMs with the exact questions your buyers ask and logging whether your brand is mentioned, cited, or recommended. This manual check is the most accurate way to see if your off-page entity footprint and on-page content are strong enough to earn AI citations.
Check 1: Query ChatGPT, Perplexity, and Gemini for Your Core Money Questions #
What to do: Identify 5 to 10 "money questions" — the high-intent questions that buyers ask when they are actively looking to hire or buy in your niche. Open ChatGPT (running GPT-5.5 or GPT-5.4 mini), Perplexity (using Sonar Pro), and Gemini (running Gemini 3.5 Flash or Gemini 3.1 Pro) in separate windows. Paste your questions verbatim and log the results.
Here is the exact prompt template I use for this check:
I am looking for a [service/product] that can help me with [specific problem] in [city/location]. Can you recommend the top 3 options? For each option, explain why they are recommended and provide a link to their website or a source verifying their work.What good looks like: Your business is recommended in the top three options across all three engines, accompanied by a direct, clickable link to your domain and a factual summary of your specific expertise.
Red flags: The AI recommends your direct competitors while completely ignoring your brand, or it recommends your brand but cites an outdated address, an incorrect service list, or a dead URL.
How to fix: You must publish highly specific, question-answering content on your own site that targets these exact queries, and ensure your off-page entity signals (directories, press, and profiles) are completely aligned.
Check 2: Audit Google AI Overviews for Your Core Search Queries #
What to do: Search your primary commercial and informational keywords in Google. If an AI Overview does not generate automatically, click the "Generate" button. Carefully examine the cited sources in the side-scrolling link carousel and the inline citation cards.
What good looks like: Your website is featured as a primary source card in the AI Overview, or your content is cited inline to support a specific factual claim in the summary.
Red flags: An AI Overview appears for your core query, but it cites low-quality aggregator sites, forums, or direct competitors while ignoring your page — even if your site ranks in the top three traditional blue links directly below the overview.
How to fix: Restructure your content using the inverted pyramid model. Place a direct, factual, 1-2 sentence answer at the very top of your page or section, and mark it up with FAQ schema. Read my playbook on why FAQ schema and AEO are the highest-impact moves for AI citation to see exactly how to structure these answers.
How Do You Run a Competitor Citation Analysis? #
You run a competitor citation analysis by extracting the exact domains cited in AI answers for your target queries and identifying the common patterns in their off-page presence. This reveals the specific directories, forums, and publications you must target to match their visibility.
Check 12: Competitor Comparison — Who IS Getting Cited for Your Queries #
What to do: Query ChatGPT, Perplexity, and Gemini for your primary commercial keywords. Ask the engine to list the sources it used to generate the answer. Log the cited domains.
Here is the exact prompt template I use for this check:
I am researching [topic/industry]. Can you provide an overview of [specific topic]? After generating the response, list every single source URL you used to synthesize this answer, and explain why those specific sources were selected as authoritative.What good looks like: Your website is cited alongside your top competitors, or you are cited as the primary authority for a specific subtopic.
Red flags: Your competitors are cited repeatedly across different engines, while your brand is completely absent, even though you have higher traditional domain authority or more backlinks.
How to fix: Analyze the cited competitor URLs. If they are cited from a specific industry directory, a forum thread, or a PR release, you must get your brand featured on those same external platforms.
Are AI Crawlers and Schema Markup Configured Correctly? #
You verify crawler and schema configuration by checking your robots.txt file for restrictive disallow rules and running your pages through Google's Rich Results Test. If AI bots are blocked from reading your site, or if they cannot find structured data to understand your entities, you will never earn citations.
Check 3: Verify AI Crawler Access in Your robots.txt File #
What to do: Open your website's robots.txt file in a browser (e.g., yourdomain.com/robots.txt). Search for rules that block major AI user-agents, such as GPTBot, PerplexityBot, ClaudeBot, and Google-Extended.
Here is an example of a robots.txt file that explicitly allows AI crawlers to access your content:
User-agent: GPTBot
Allow: /
User-agent: PerplexityBot
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: Google-Extended
Allow: /What good looks like: Your robots.txt file explicitly allows these bots, or at least does not block them with a blanket Disallow: / under User-agent: *.
Red flags: A blanket Disallow: / rule that blocks all crawlers, or specific blocks on GPTBot or PerplexityBot that prevent them from crawling your commercial pages.
How to fix: Update your robots.txt file to ensure AI crawlers have full access to your public-facing pages. If you want to block AI engines from training on your content but still want them to cite you, allow them to crawl your site but use opt-out settings on the LLM platforms directly where possible.
Check 4: Validate Your Schema Markup with Google's Rich Results Test #
What to do: Open the Schema Markup Validator (validator.schema.org) or Google's Rich Results Test. Enter your homepage and key service URLs to check for valid Organization, LocalBusiness, and FAQPage schema.
Here is an example of a clean, valid LocalBusiness JSON-LD schema:
{
"@context": "https://schema.org",
"@type": "LocalBusiness",
"name": "William Spurlock Studio",
"url": "https://williamspurlock.com",
"logo": "https://williamspurlock.com/logo.png",
"address": {
"@type": "PostalAddress",
"streetAddress": "123 Main Street",
"addressLocality": "New York",
"addressRegion": "NY",
"postalCode": "10001",
"addressCountry": "US"
}
}What good looks like: Your pages return zero errors and zero warnings for these schema types, with fully resolved entity properties like name, url, sameAs, and logo.
Red flags: Missing structured data entirely, or schema that contains broken syntax, missing required fields, or unresolvable URLs.
How to fix: Implement clean, valid JSON-LD schema on your site. Ensure your sameAs array links to your official social profiles, Wikidata, and Google Business Profile to help AI engines connect your off-page entities.
How Do You Track AI Bot Hits in Server Logs? #
You track AI bot hits by analyzing your raw server logs or Cloudflare analytics for requests matching the user-agent strings of major AI crawlers. This log analysis provides direct, empirical proof of which pages AI engines are actively indexing.
Check 10: Check Server Logs and Cloudflare Analytics for AI Bot Hits #
What to do: Log into your Cloudflare dashboard or access your raw Nginx/Apache server logs. Filter requests by User-Agent, looking specifically for GPTBot, PerplexityBot, ClaudeBot, and Google-Extended.
What good looks like: Frequent hits from AI crawlers on your high-value blog posts and service pages, showing that the engines are regularly updating their index of your brand facts.
Red flags: Zero hits from AI crawlers over a 30-day period, or crawlers hitting only your homepage while ignoring your deep informational content.
How to fix: Submit your sitemap directly to search engines, ensure internal linking is clean, and verify that your server is not returning 403 or 429 status codes to legitimate AI user-agents.
Does Your Brand Exist as a Trusted Entity on Wikipedia and Wikidata? #
Your brand exists as a trusted entity if it has a verified Wikidata item, a Wikipedia page, or a Google Knowledge Panel that establishes your business in the global knowledge graph. AI engines rely on these structured knowledge bases to verify facts and resolve entity relationships.
Check 7: Verify Wikipedia, Wikidata, and Knowledge Panel Entity Presence #
What to do: Search your brand name on Wikidata (wikidata.org) and Wikipedia. Search your brand name on Google to see if a Knowledge Panel appears on the right side of the search results.
What good looks like: A fully populated Wikidata item containing your official website, founder, and founding date, and a Google Knowledge Panel that correctly attributes your brand facts.
Red flags: No entity presence in Wikidata, or a Knowledge Panel that confuses your business with a completely unrelated company or person.
How to fix: Create a Wikidata item for your business using verifiable sources, such as official state business registries or high-authority press releases.
Check 11: Audit NAP and Brand Fact Consistency Across the Web #
What to do: Search your business name, address, and phone number (NAP) on Google. Compare the details on your website with those on Yelp, Facebook, Apple Maps, and other major directories.
What good looks like: Identical NAP details and brand facts (such as founding year and service list) across every single platform.
Red flags: Variations in your business name (e.g., "William Spurlock Studio" vs. "Spurlock Web Design"), old phone numbers, or different addresses on different directories.
How to fix: Use a local citation service or manually update your profiles to ensure absolute consistency. AI engines are highly sensitive to conflicting facts and will skip citing businesses with inconsistent data.
How Do Reddit and Google Business Profile Drive AI Citations? #
Reddit and Google Business Profile drive AI citations by providing high-authority, user-generated reviews and real-time local signals that AI engines prioritize for commercial recommendations. ChatGPT and Perplexity frequently crawl Reddit to find authentic user opinions, while Google AI Overviews extract local business facts directly from Google Business Profile.
Check 5: Audit Your Google Business Profile and Review Footprint #
What to do: Log into your Google Business Profile dashboard. Check your review volume, average rating, and the frequency of your responses to reviews.
What good looks like: A 4.5+ rating with dozens of detailed reviews that mention your specific services, and regular owner responses to every review.
Red flags: An unclaimed profile, low review volume, or reviews that contain negative keywords that AI engines might extract as warning signals.
How to fix: Claim your profile, optimize your categories, and implement a systematic review generation process to build a steady stream of fresh, keyword-rich reviews.
Check 6: Search Your Brand on Reddit and Forums #
What to do: Search Google using site operators: site:reddit.com "your brand name" or site:reddit.com "your industry" "recommendation". Log whether users are naturally recommending your services in relevant threads.
What good looks like: Active, positive discussions where real users recommend your business by name and link to your website.
Red flags: Zero mentions of your brand on Reddit, or negative threads where users warn others against using your services.
How to fix: Participate genuinely in relevant subreddits by answering questions and sharing expertise, without spamming your links. AI engines use Reddit as a primary source of truth for unbiased user recommendations.
Is Your Content Structured and Fresh Enough for AI Extraction? #
Your content is structured and fresh enough if it uses question-based H2 headings, opens with bold direct answers, and includes active lastModified timestamps in the metadata. AI engines prioritize content that is easy to extract and has been updated within the last six months.
Check 8: Test the Answer-Extractability of Your Key Pages #
What to do: Copy a 500-word section of your key service page. Paste it into ChatGPT or Claude and ask: "Extract the direct answer to [your target question] from this text in under 50 words."
What good looks like: The AI immediately extracts a clean, accurate, and complete answer that matches your brand's core value proposition.
Red flags: The AI returns a vague summary, misses the main point, or states that the text does not contain a direct answer.
How to fix: Rewrite your content using the inverted pyramid model. Place a bold, direct answer immediately under your H2 question heading, and follow it with supporting bullet points.
Check 9: Audit Content Freshness and lastModified Timestamps #
What to do: Inspect your page source or sitemap to verify the presence of datePublished and dateModified schema. Check the actual publication and update dates of your high-value blog posts.
What good looks like: Every informational page has been updated within the last six months, with an active lastModified timestamp that matches the actual update date.
Red flags: Blog posts that show publication dates from several years ago with no modification updates, or sitemaps that lack lastmod tags entirely.
How to fix: Implement a content refresh schedule. Update outdated statistics, add fresh insights, and ensure your sitemap and schema automatically update the dateModified field.
How Do You Score Your AI Visibility? #
You score your AI visibility by assigning zero to two points for each of the twelve checks and summing them to find your overall readiness tier. This simple scorecard lets you quickly identify whether your business is highly visible, moderately exposed, or completely invisible to AI engines.
Use the table below to score each of the twelve checks:
| Check | Area | 0 Points | 1 Point | 2 Points |
|---|---|---|---|---|
| 1 | ChatGPT/Perplexity/Gemini | Completely ignored | Mentioned without links | Top 3 recommendation + link |
| 2 | Google AI Overviews | Competitors cited only | Mentioned in text only | Featured source card + inline link |
| 3 | AI Crawler Access | Blocked in robots.txt | No explicit block but no allow | Explicitly allowed |
| 4 | Schema Markup | No schema markup | Valid schema with warnings | Valid schema, zero errors/warnings |
| 5 | Google Business Profile | Unclaimed or inactive | Claimed but low review volume | 4.5+ rating, active responses |
| 6 | Reddit and Forums | Zero mentions | Mentioned but no links/praise | Active positive recommendations |
| 7 | Wikipedia and Wikidata | No entity presence | Knowledge Panel but no Wikidata | Wikidata item + Knowledge Panel |
| 8 | Answer-Extractability | AI cannot extract answer | AI extracts vague summary | AI extracts clean direct answer |
| 9 | Content Freshness | Updated > 12 months ago | Updated 6-12 months ago | Updated < 6 months ago |
| 10 | AI Bot Hits in Logs | Zero hits in 30 days | Rare hits on homepage only | Frequent hits on deep pages |
| 11 | NAP Fact Consistency | Conflicting facts on web | Mostly consistent with minor typos | 100% consistent NAP everywhere |
| 12 | Competitor Comparison | Competitors own all citations | Competitors share citations | Your brand owns dominant share |
Interpret Your Score #
Sum your points across all twelve checks to find your readiness tier:
- 0-8 Points: AI Invisible. Your business is completely hidden from answer engines. AI crawlers are likely blocked, schema is missing, and your brand facts are inconsistent. You are losing high-intent leads to competitors every day.
- 9-16 Points: AI Exposed. You have some baseline visibility, but you are not winning the primary citations. Your technical foundation is decent, but your content structure and off-page entity signals are weak.
- 17-24 Points: AI Dominant. Your business is fully optimized for the AI era. You are winning high-value citations across ChatGPT, Perplexity, and Google AI Overviews.
Frequently Asked Questions About AI Visibility Audits #
How often should I run an AI visibility audit? #
You should run an AI visibility audit at least once every quarter to keep up with fast-moving model updates and crawler behaviors. Because search engines like Google update their AI Overview extraction algorithms frequently, a quarterly audit ensures your content structure remains aligned with the latest standards.
Does blocking AI crawlers in robots.txt hurt my SEO? #
Blocking AI crawlers like GPTBot does not hurt your traditional Google rankings, but it completely prevents your site from being cited in ChatGPT and Perplexity. If you disallow these bots, you are opting out of the massive search volume shifting to answer engines.
What is the most important schema markup for AI visibility? #
The most important schema markup for AI visibility is LocalBusiness or Organization schema combined with FAQPage JSON-LD. These structured data types establish your brand as a verifiable entity in the knowledge graph, making it 50% easier for AI engines to extract and cite your facts.
Can I pay to get my business cited in ChatGPT or Perplexity? #
No, you cannot pay for organic citations in ChatGPT or Perplexity, as their recommendations are generated purely through algorithmic extraction of authoritative web sources. Winning these citations requires building real topical authority and structured on-page content.
Why does Google AI Overview show different sources than the traditional blue links? #
Google AI Overviews use a separate retrieval-augmented generation (RAG) pipeline that prioritizes information density and direct answers over traditional backlink authority. This is why a site ranking at position eight in the blue links can easily win the primary AI Overview citation if its content is better structured for extraction.
How do I track Perplexity traffic to my website? #
You can track Perplexity traffic by analyzing your referral traffic in Google Analytics, where it appears under the hostname perplexity.ai. Monitoring this referral traffic alongside your server log hits from PerplexityBot gives you a clear picture of your answer engine referral volume.
What are the user-agent strings for the main AI crawlers? #
The primary user-agent strings you need to track are GPTBot for OpenAI, PerplexityBot for Perplexity, ClaudeBot for Anthropic, and Google-Extended for Google's AI models. Ensuring these crawlers are allowed in your robots.txt file is the first step to earning citations.
How does Reddit affect my business's AI visibility? #
Reddit has a massive impact on your AI visibility because ChatGPT and Perplexity have direct licensing agreements to crawl and prioritize Reddit threads for subjective recommendation queries. If real users are not recommending your brand on Reddit, AI engines are highly unlikely to recommend you to buyers asking for the "best" options.
Ready to Dominate the Answer Engines? #
If you want to stop losing traffic to AI-generated answers, you must transition from traditional SEO to a dedicated answer engine optimization strategy. While a DIY audit is a great first step, implementing a comprehensive AEO framework requires deep technical and structural expertise.
I help business owners and premium brands audit, structure, and optimize their digital footprint to win dominant citations in ChatGPT, Perplexity, and Google AI Overviews. If you want a professional, done-for-you AI visibility audit that identifies every crawler block, validates your schema, and restructures your key pages for maximum extraction, let's talk.
Book an AI visibility audit today to secure your brand's presence in the AI era.
Related Posts

Declaring Independence From Google: The Traffic Diversification Playbook
Relying on Google for 70%+ of your traffic is a high-risk gamble in 2026. Here is the five-channel playbook to diversify your pipeline and make Google optional.

Entity SEO: How to Make Your Brand Something AI Actually Knows
AI search engines don't rank page links; they reason about brand entities. Learn how to establish your brand node in the knowledge graphs ChatGPT and Perplexity trust.

llms.txt, robots.txt, and AI Crawlers: Who Should Be Allowed to Read Your Site?
Blocking all AI crawlers out of panic is killing your brand's AI visibility. Learn how to configure robots.txt and llms.txt to block training bots while allowing search bots.
