Most brands know their Google rankings. Far fewer know whether ChatGPT, Perplexity, or Google's AI Overviews recommend them at all — and that gap is easy to miss, because nothing tells you it exists until a customer mentions it.
An AI visibility audit closes that gap. It answers questions like:
- When someone asks ChatGPT, Perplexity, or Google AI about our category, does our brand show up?
- What topics is our brand getting recommended on, and where are we missing?
- Is our current website health and content strategy set up to help us succeed in AI search?
This guide shows you how to run that audit using Semrush's AI Visibility Toolkit.
You'll measure where your brand appears across AI platforms, which content gets cited, and where topical gaps exist. Any marketer with Semrush access can run an AI visibility audit alone — it doesn't require an agency or a dedicated data team — and the process scales down for smaller businesses too.
Don't have Semrush yet? Try our free AI visibility checker to get started without a commitment, or read our lighter-weight walkthrough if you want a faster starting point.
What is an AI visibility audit?
An AI visibility audit is a structured review of whether your brand is recommended in AI-generated answers — ChatGPT responses, Google AI Overviews, Perplexity summaries — for the prompts and searches most relevant to your category.
AI visibility audits ask a different question than the ranking checks you already run: not just whether you rank, but whether AI systems trust your brand enough to name it. That depends on different inputs — crawlable and extractable content, consistent entity signals, and a track record other sources corroborate — which is what the rest of this audit measures.
How to run an AI visibility audit
Run an AI visibility audit by seeing how you currently show up in AI answers and where you’re missing from the conversation. Do that in six steps:
- Establish your baseline for AI visibility
- Identify your citation-worthy content
- Find topics and prompts where your brand is missing
- Analyze brand sentiment and the AI narrative
- Identify influential off-site sources shaping AI answers
- Check site-level issues that affect AI access
Let’s explore each step.
Step 1: Establish your baseline for AI visibility
Establishing a baseline for AI visibility shows you if AI systems reference your brand name or cite your content as a source more than competitors.
Start by opening AI Visibility Overview in the AI Visibility Toolkit to see how your brand appears in AI answers.
This dashboard shows mention counts, cited pages, and distribution across AI platforms (ChatGPT, Google AI Overviews, AI Mode, Gemini).
Write down your brand's AI Visibility score, mentions, and cited pages. Use these high-level metrics to benchmark your site against your competitors.
For Allbirds, we see 4.9K mentions and 1.9K cited pages.
Compare this baseline against competitors using the Competitor Research tool to see how that baseline stacks up against your closest competitors. For a deeper walkthrough, see our guide to benchmarking brand mentions in AI answers.
Step 2: Identify your citation-worthy content
Your citation-worthy content — the pages already earning AI citations — tells you what to publish and optimize next.
Navigate to the "Cited Pages" section in Visibility Overview.
This shows which URLs from the domain are referenced in AI responses and how many prompts from the Semrush database each page currently gets cited for.
In the Allbirds example, educational and comparison pages earn more citations than product or category pages. That's a signal to optimize the educational pages already earning citations and create more like them.
Next, expand a page to see the prompts and whether the brand name is "Mentioned" or "Missed" within those answers that cite the page.
- Cited and mentioned: A visibility win
- Cited but not mentioned: An opportunity to update the page and get mentioned for this prompt, as long as other brands are getting mentioned
Compare AI cited pages to SEO Top Pages
Comparing AI-cited pages to your SEO Top Pages shows you which pages rank well in search but earn no AI citations — a clear list of high-traffic pages worth optimizing for citations too.
You can use the Top Pages report to compare this to the pages that attract the most SEO traffic. Notice in the case of Allbirds, the organic traffic is mostly driven by collection and category pages.
These pages already rank well in organic search, but that's not the same as being AI-citable: Collection and category pages are built for browsing, not for the concise, self-contained answers AI systems extract and cite.
Now, optimize them for AI retrieval — add clear structure, use question-and-answer formats, and concise summaries that AI systems can extract and cite.
Step 3: Find topics and prompts where your brand is missing
Finding topics and prompts where your brand is missing (called AI visibility gaps) reveals where AI recommends your competitors instead of you.
In Visibility Overview, scroll to "Topics & Sources" and select "Topic Opportunities."
Each topic shows where AI-generated answers exist and attract demand.
For Allbirds, high-volume topics like "Running Shoes" and "Diabetic and Neuropathy Shoes" show their competitors mentioned, but not Allbirds. Each topic can be expanded to reveal the specific prompts generating those answers.
Take note of the missed prompts to identify questions that your existing content doesn't answer. Then, create content that addresses those questions clearly and directly.To find more related topics and prompts, along with their AI volume and relevance, use Prompt Research.
Step 4: Analyze brand sentiment and the AI narrative
Analyzing brand sentiment and the AI narrative helps you see which brands are framed positively, neutrally, or negatively.
Unlike search results, AI doesn't rank brands by their URL. It can, however, describe a brand as an option with pros and cons alongside other options.
Use the Brand Performance reports to review sentiment and share of voice compared to your rivals.
In this snapshot, Allbirds is usually framed favorably (88%) and holds the largest share of voice (14%) in AI answers compared to the competition.
You can navigate to the “Key Sentiment Drivers” table in the Perception Report to review the brand strength factors and areas for improvement.
Step 5: Identify influential off-site sources shaping AI answers
Identifying influential off-site sources shaping AI answers lets you find which sources have influence in your niche, so you can pinpoint where your brand needs to appear.
AI-generated answers often cite third-party user generated content (UGC) platforms like Reddit, YouTube, Quora, LinkedIn, and Medium. Back in the Visibility Overview, check "Cited Sources" to see which external domains are most frequently cited across your core topics.
The domains that appear most often are shaping how AI systems explain your category and which brands get named, so they're where your brand needs to show up.
You can use this data to determine if it's worthwhile to invest in growing your brand on YouTube or managing your reputation on Reddit.
Step 6: Check site-level issues that affect AI access
Check for site-level issues that affect AI access so you can fix any errors that could be hurting your visibility.
Run a quick scan with Site Audit's free check, then go deeper with the full site-wide audit covering issues like the ones listed below. The "AI Search Health" feature checks structured data and AI crawler access directly.
Common issues affecting AI visibility:
- Weak internal context: Links without descriptive anchor text reduce how clearly pages are framed for AI systems. Internal links also signal to AI systems which pages belong to which topics, so pages without that context are less likely to be associated with your core subject matter.
- Structurally isolated pages: Pages with only one incoming internal link are less likely to be referenced consistently
- Missing AI guidance file: A missing llms.txt doesn't block access, but removes a clear signal about how AI should interpret site content
These issues don't stop AI systems from crawling your pages. They make most of those pages harder to use as sources, so AI cites only the handful that are clear enough. Fixing them makes more pages easier for AI systems to use as sources.
How to deliver your AI visibility audit
Deliver your AI visibility audit as a prioritized list of gaps, baseline benchmarks against competitors, and a clear next step for each.
The most useful audits reveal patterns: where AI consistently includes or excludes a brand, which content types AI relies on, and how similar prompts are handled across the category.
Here are three options for delivery:
- Document illustrated with screenshots: Combine key Semrush screenshots with brief comments of what each report shows and what it implies. Quick to generate and easy to share by email.
- Standardized PDFs with Semrush My Reports: Use Semrush's My Reports templates to consolidate metrics in one branded format. Connect Google Analytics to include widgets for analyzing AI referral traffic. These are ideal for recurring client reports.
- Video walkthrough: Pair a written audit with a Loom recording or live presentation to explain how findings connect, highlight priorities, and clarify tradeoffs. Best when alignment is needed before action.
Choose the format based on how much context your recipient needs to decide what to do next.
What to do after your AI visibility audit
After your AI visibility audit, you'll want to fix technical issues, improve citation readiness, and expand topic coverage.
The roadmap for each of these will be unique for each business and situation, but here's a suggested starting point.
- Immediate (week 1): Fix technical issues that limit AI access. Resolve problems affecting whether AI systems can access, interpret, or attribute your content. These fixes have site-wide impact and set the foundation for everything that follows.
- Near term (weeks 2-4): Improve citation readiness. Focus on improving content that's not already cited. Make existing pages clearer, more complete, and easier for AI systems to reuse. Set up Prompt Tracking on your highest-priority prompts to monitor progress continuously, rather than waiting to manually re-run the full audit.
- Mid term (months 2-3): Expand topic coverage. Address the topical gaps by building new content in those areas where your brand is missing.
Turn AI visibility audit findings into clear priorities
As AI systems increasingly shape how options are compared and recommended, many brands still lack visibility into why their brand appears in some answers and not others.
With Semrush, AI search patterns become observable signals you can review, prioritize, and revisit over time. Run your AI visibility audit with Semrush before your competitors do, and stay ahead in your market.
