How To Measure Your Brand’s Visibility In AI Answers (And Fix What You Find)

This post was sponsored by Hubspot. The opinions expressed in this article are the sponsor’s own.

Why is my page ranking on page one but not appearing in ChatGPT answers?

How do I know whether ChatGPT, Claude, and Perplexity are recommending my brand?

What do I change to show up for the prompts that lead to a sale?

If you’ve asked these questions, you’ve probably typed your own brand into ChatGPT to see what came back.

Unfortunately, a handful of AI visibility spot checks will not tell you whether your brand shows up in AI search 30% of the time or 80% of the time.

Why Does My Page Rank On Page One While ChatGPT Recommends Someone Else?

Traditional SERP rankings and AI search recommendations run on separate systems.

Traditional SEO rankings measure how your page performs against a query on a results page, evaluating factors such as keyword relevance, backlink profiles, technical site structure, and user engagement signals to determine your URL’s position against other sites in traditional search engine list listings.

AI search tools, like ChatGPT, Gemini, and Perplexity, write a unique, new answer each time someone asks a question. Common AI answers are made by each LLM assembling the AI answer from exterior sources that it treats as authoritative on the topic. This logic-type seems to be the primary differentiator between AI search visibility and SERP rankings.

This helps answer one of the most common questions we hear, “Why is my website losing traffic despite good SEO rankings?” The new SERP phenomenon of increased conversions and lower traffic stems from your traditional visitor getting everything they need from a synthesized version of your website, via AI answer.

There’s still good news, though. Whenever your brand does get named, through a mention, citation, or recommendation, the buyer still takes the information forward, even if you don’t get the official visit.

What’s The Difference Between Answer Engine Optimization & Traditional SEO?

Traditional SEO is the practice of getting your website onto search results pages and search engines make ranking decisions based on the properties you own: your pages, your internal linking, your Core Web Vitals, your schema, with validation from backlinks.

Answer engine optimization, or AEO, is the practice of getting your brand mentioned, cited, and recommended accurately inside the answers ChatGPT, Claude, Gemini, and Perplexity generate.

AEO does not replace SEO. Each LLM still has to crawl and read your site before it can use you as a source, so your technical health, site structure, schema markup, and content quality remain the foundation everything else sits on.

The practical difference is coverage across the web that further informs the recommended websites in the answer.

Both your source mix, which is the exact websites that contribute information about your brand to AI engines, and your visibility score come out of the same audit.

How Do I Know If AI Search Engines Like ChatGPT Are Recommending My Brand?

This can be identified by reading the AI answer. If you have been recommended by AI search, that means you were listed among the options the buyer should actually consider.

A mention is different. Your brand gets named somewhere in the answer without being put forward as an option.

A brand can be mentioned in most answers and recommended in almost none. The recommendation number is the one that tracks with pipeline, which is why the audit below counts both.

How To Run Your First AI Visibility Audit

This is where another question we hear constantly gets answered: “How can I track whether my brand appears in AI search results?”

One pass through the steps below produces: 

Phase 1: Turn Your Ranked Keywords Into AI Search Prompts

Step 1: Export your commercial-intent queries from Search Console.

1. Open the Performance report, set the date range to the last three months, and pull the queries where your average position is 10 or better.
2. Then narrow to commercial intent: terms containing “best,” “software,” “tool,” “platform,” “vs,” “alternative,” “pricing,” or your category name.

Step 2: Build your prompt tracking list.

1. Rewrite each SEO search query as a buyer would ask it in real life.
2. Add the constraints a real buyer includes to make it read more like a hyperspecific long-tail keyword: company size, industry, budget, or the specific job they need done.“Best dog park near me” becomes “What is the best dog park around 01002 that has enough play space for two Siberian huskies?”
3. Check that each rewrite reads as a full question rather than a keyword string.
4. Place these rewrites into a spreadsheet for tracking.

Pro Tip: HubSpot AEO runs Prompt Tracking across ChatGPT, Gemini, and Perplexity on a continuous basis, so your list updates without a manual rebuild each quarter.

Phase 2: Collect Your AI Answer Data

Step 3. Run every newly-created AI prompt on every engine.

1. Sign out, open a temporary chat, or use incognito mode so account history and saved memory do not shape the answer.
2. Run each prompt two or three times.
3. Record the answer that appears most often.
4. Repeat on every engine.

Step 4. Notate 6 fields for every prompt.

1. Which engine returned each answer.
2. Which brands are mentioned.
3. The order those brands appear in.
4. Whether your brand is mentioned at all.
5. Whether your brand appears among the first three named.
6. The domains cited in the answer.

Pro Tip: HubSpot AEO’s Citation Analysis groups the citations behind your tracked prompts by domain and by content type, identifying which of those source types are feeding answers in your category.

Phase 3: Find The Queries Where You Rank But Do Not Get Cited 

Step 5. Calculate your AI Visibility Score for each engine separately.

1. Divide mentions by prompts tested to get your mention rate.
2. Divide top-three appearances by prompts tested to get your recommendation rate.
3. Divide your mentions by all brand mentions returned to get your share of voice.
4. Average the three engines for a composite, and report the per-engine figures alongside it.

That math answers the other question we field constantly: “How can I measure my brand’s visibility in generative AI search responses?”

Hate math? Here’s a free AI Search Grader to get you started.

Step 6. Pull your gap list and your source mix from the same log.

1. Flag every prompt where you hold a top-10 ranking and your brand is not mentioned. That is your gap list.
2. Tag each cited domain against the six source types, then total the tags. That is your source mix.

The most common sources that AI cites include: 

• Review directories such as G2, Capterra, and TrustRadius
• Online communities, meaning Reddit threads and category forums
• Third-party roundup articles, the “best X software” listicles
• News and press coverage
• Reference pages such as Wikipedia
• Your own website

Pro Tip: HubSpot AEO runs your prompt set on a schedule and reports Brand Visibility Score, Share of Voice against named competitors, and sentiment across all three engines, which lets you see who AI is recommending in your category without maintaining the spreadsheet.

What Do I Change To Show Up For The Prompts That Lead To A Sale?

Now that you know which commercial or transactional intent prompts you show up for in AI search, here is how to approach fixing the source material that leads to the AI answers returned for your prompts. 

How To Fix Your Presence On These Sources, or Commonly-Cited Websites

Fixing your presence on the websites below help keep AI answers true to your brand and lead to more accurate citations and recommendations. 

Review directories come first because they get cited most often on commercial prompts, since those pages are already structured as comparisons with features, pricing, ratings, and competitor sets in one place. Your own website comes last because it is usually the smallest share of any source mix, even though it is the only one you fully control.

If your own source mix ranked them differently, follow your data instead.

Review Directories

1. Fill out every field in your G2, Capterra, and TrustRadius profiles.
2. Confirm you are listed in the categories your buyers actually browse.
3. Keep new reviews arriving rather than letting a batch age.

Online Communities

1. Use your source mix log to find the exact Reddit threads and forum posts that were cited.
2. Answer questions in those threads under a real identity.
3. Correct outdated or wrong claims about your product.
4. Skip the threads where you have nothing useful to add.

Third-Party Roundup Articles

1. Use your source mix log to identify which “best X software” listicles are cited for your prompts.
2. Check whether your existing entries on those pages are out of date.
3. Approach the publisher with accurate current specs.

Your Own Website

1. Lead with the direct answer in the first sentence rather than building to it.
2. State capabilities in plain declarative sentences an LLM can lift.
3. Keep a comparison page that is accurate about what you do and do not do.

Pro Tip: Rather than ranking the work yourself, HubSpot AEO returns AEO Recommendations tied to the specific prompts you are losing, which lets you find out which sites AI cites in your category and sequence the work in the same session.

The Playbook HubSpot Ran On Itself First

HubSpot’s marketing team built this method internally before the product existed, working the same source order this guide lays out.

Rebuilding their software comparison content drove a 642% increase in citations. Rebuilding FAQ and glossary pages produced a 60% increase in citation share on related prompts. Their Reddit citations went from 178 in May 2025 to roughly 146,000 by December 2025.

Across the program, that produced a 433% increase in citations, an 1,850% increase in qualified leads from AI, and AI-sourced leads converting at 3x the rate of leads from other channels.

Now, you can give it a try by hand or access the tool for free for the first 28-days, then, only $50 per month, with no other HubSpot subscription required.

Frequently Asked Questions

What are the best tools for answer engine optimization and AI search visibility?

The right tool depends on how many engines you need covered and whether you want reporting or a work order. Check engine coverage at the entry price, prompt-level rather than keyword tracking, and whether recommendations are included. HubSpot AEO covers ChatGPT, Gemini, and Perplexity at $50 per month with prioritized recommendations.

What should I look for in an AI visibility tool?

Five things to check before you commit to any platform:

• How many engines are covered at the entry price. Several tools track one engine at the entry tier and put the rest behind an upgrade, which matters because your per-engine numbers will diverge.
• Whether it tracks prompts or keywords. Buyers ask questions with constraints attached, and those do not map cleanly onto keywords.
• Whether it reports citations or only mentions. A mention count gives you the score. The cited domain tells you which site to work on.
• Whether it returns recommended actions. Reporting gives you a baseline. Recommendations give you a work order.
• What multi-engine coverage actually costs, rather than the headline entry price.

What platforms help track brand mentions and citations in AI search results?

Mentions and citations are two different measurements. A mention is your brand named in the answer. A citation is a page the model drew from to build it. HubSpot AEO reports both, grouping citations by domain and content type so you can trace a mention back to the review site, forum thread, or owned page that produced it.

What are the top solutions for monitoring AI search positioning?

Positioning means where your brand lands inside an answer, not only whether it appears, since a brand named first carries more weight than one named fourth. HubSpot AEO reports position within the answer alongside share of voice, tracked over time so you can see month-over-month movement.

What tools help optimize content for generative engine optimization?

The useful ones connect citation data to a specific action. HubSpot AEO’s AEO Recommendations identify which pages and off-site sources are driving citations for your tracked prompts, then prioritize what to create or update against the prompts you are currently losing.

What are the best AI SEO platforms for tracking zero-click search performance?

Zero-click performance in AI search is measured through mentions and citations rather than sessions, because the answer resolves without a visit. HubSpot AEO measures mention frequency, position within the answer, and share of voice across ChatGPT, Gemini, and Perplexity, none of which requires a click.

What Does HubSpot AEO Track?

Your mention rate and recommendation rate tell you where you stand. What moves them is the set of sites the engines cite.

How much does HubSpot AEO cost, and is there a free trial?

HubSpot AEO is $50 per month for 25 tracked prompts across ChatGPT, Gemini, and Perplexity, with a lower annual rate. A free 28-day trial covers the same 25 prompts with no credit card required, and you can add prompts at any time.

Where can I find HubSpot AEO reviews on G2, Capterra, and TrustRadius?

HubSpot maintains listings on the major B2B review directories, with AEO reviews appearing under its software category pages. Those directories are also among the sources AI engines cite most on commercial prompts.

Image Credits

Featured Image: Image by Hubspot Used with permission.