How to Test If ChatGPT and Perplexity Can Find Your Business
Only 11% of domains cited by ChatGPT are also cited by Perplexity. That means being visible on one AI platform does not protect you on the others. And most businesses are not visible on any of them.
Skip the manual testing. Get your free Blind Spot Report and see exactly how you appear across ChatGPT, Perplexity, Gemini, and Google AI right now.
In This Article
- Why the Assumption of AI Visibility Is Almost Always Wrong
- The 4-Platform AI Visibility Test
- What to Look For in AI Responses
- Mentioned vs. Recommended: A Critical Distinction
- Why Each Platform Sees You Differently
- What Complete Invisibility Is Telling You
- Scoring Your AI Visibility Results
- Frequently Asked Questions
Why the Assumption of AI Visibility Is Almost Always Wrong
When business owners are asked whether they show up on ChatGPT or Perplexity, most say yes or assume yes. When they actually check, the majority are not there.
The confusion is understandable. Most businesses have been working on their online presence for years: a website, a Google Business Profile, some reviews. It feels like that work should count. And it does, partially, for traditional search. But AI citation operates on different logic. A business can be on the first page of Google for its primary keyword and still be completely invisible when a customer asks ChatGPT for a recommendation.
The test below takes 20 minutes and will tell you more about your AI visibility than any analytics dashboard.
Do Not Test with Your Business Name
Searching for your own business name tells you almost nothing. What matters is whether AI recommends your business when a customer searches for your service type and location. Test with the actual queries your potential customers would use, not the queries you already know you appear in.
The 4-Platform AI Visibility Test
Open four browser tabs simultaneously and run identical queries on each platform. Use natural language questions the way a customer would actually phrase them, not keyword-style searches.
| Platform | Where to Test | What Makes It Different | Query Format to Use |
|---|---|---|---|
| ChatGPT | chat.openai.com | Draws from training data + Bing web search. High weight on entity recognition. | "Who is the best [type] in [city]?" |
| Perplexity | perplexity.ai | Live web search with heavy freshness weighting. Cites sources explicitly. | "Recommend a [type] near [city]" |
| Google Gemini | gemini.google.com | Draws heavily from Google index, GBP data, and structured data. | "Find me a [type] in [city]" |
| Google AI Overview | google.com search | Appears above organic results for many local queries. GBP-weighted. | Search: "[type] in [city]" |
Run at least three different query variations per platform. A single test can produce an outlier result. You want to know the pattern across queries, not whether you appear in any single response.
Prefer a professional analysis? Call (213) 444-2229 to talk through what our diagnostic process covers across all four platforms.
What to Look For in AI Responses
When you run your test queries, record exactly what happens in each response. AI visibility is not binary. There are four distinct outcome states, and each tells you something different about your citation standing.
Document every result in a simple spreadsheet: platform, query, outcome state, which businesses appeared instead of you. This becomes the baseline you measure improvement against.
Mentioned vs. Recommended: A Critical Distinction
Business owners who appear in AI responses often celebrate without noticing the difference between a mention and a recommendation. The distinction is worth understanding because they have very different conversion implications.
An AI Recommendation Looks Like:
- "For plumbing in Glendale, I would recommend Martinez Plumbing. They specialize in emergency repairs and have strong reviews for fast response times."
- "Based on my search, Rivera Dental in Pasadena is well-reviewed for family dentistry and is currently accepting new patients."
- "If you need a personal injury attorney in Los Angeles, consider Nguyen Law Group, which has a track record of favorable settlements."
An AI Mention Looks Like:
- "There are several plumbers in the Glendale area including Martinez Plumbing, Acme Pipes, and Valley Plumbing Services."
- "You might want to search for local dentists in Pasadena. Some options include Rivera Dental, Smile Studio, and Pasadena Family Dentistry."
- "I found a few personal injury attorneys in the area." [without naming or distinguishing yours]
Recommendations come with implicit endorsement. Mentions are just lists. The business that gets recommended converts at a much higher rate than the business that gets listed alongside five competitors with no differentiation.
Stuck at "mentioned" and want to move to "recommended"? Get your free Blind Spot Report to find out what is blocking the upgrade.
Why Each Platform Sees You Differently
The most counterintuitive finding in AI citation research is how little the platforms agree with each other. That 11% overlap figure means businesses that have achieved visibility on ChatGPT are almost certainly invisible on Perplexity, and vice versa.
This happens because each platform draws from different source hierarchies and weights signals differently.
A business that optimizes only for Google AI Overviews by focusing on GBP completeness may remain invisible on ChatGPT, which weighs different signals. For deeper analysis on how ChatGPT and Perplexity differ specifically, see our comparison of how Perplexity vs. ChatGPT grounding differs.
What Complete Invisibility Is Telling You
If you run the 4-platform test and your business does not appear on any platform across multiple queries, that is a diagnostic signal, not just a bad outcome. It points to one of several root causes.
Common Root Causes of Complete AI Invisibility
Entity ambiguity: your business name, address, or category is inconsistent across sources. Review deficit: your review profile is too thin or too stale for AI to treat you as a trusted source. Content extractability failure: your website does not have clear, quotable statements about your service type and location. Schema absence: no structured data signals your business identity to AI crawlers. Citation isolation: no third-party sources validate your existence beyond your own website.
Complete invisibility is actually easier to diagnose than partial visibility. When you are completely absent, the root cause is almost always a foundational signal gap rather than a nuanced ranking factor. Addressing the core gaps tends to produce visible results within 4 to 8 weeks.
For the most common patterns in businesses that fail the AI citation check, our detailed analysis of why businesses are not showing up in AI search covers each failure pattern and what typically causes it.
Scoring Your AI Visibility Results
After running your 4-platform test, score your results using the framework below. This gives you a baseline visibility score and helps you identify which platforms represent the biggest opportunity.
AI Visibility Scoring Framework
| Outcome State | Score per Platform | What It Means |
|---|---|---|
| Actively Recommended | 25 points | Strong citation standing. Defend and expand. |
| Passively Mentioned | 15 points | Partial visibility. Trust and specificity gaps present. |
| Categorically Present | 5 points | Entity recognized but insufficient for citation. Competitors dominating. |
| Completely Invisible | 0 points | Foundational signal gaps. Entity record needs rebuilding. |
The Value of the Baseline
The test above gives you a snapshot. What matters is measuring the same test monthly and watching the score improve as you address visibility gaps. Businesses that start at 0-40 and reach 80+ within 6 months see measurable lead flow increases from AI channels.
For a faster, structured version of this audit, our 5-minute AI visibility audit walks through the core diagnostic questions in a quick self-assessment format.
Not sure how to interpret what you found? Send us your test results and we will tell you exactly what they mean for your business category.
Benchmark Your Results Against Your Competitors
Your AI visibility score only tells half the story. The other half is how your score compares to the competitors AI is currently recommending instead of you. A score of 40 is a problem if your top competitor scores 90. It is less urgent if the entire local market scores under 30 and the category is generally underserved in AI search.
After you run your 4-platform test and record your results, run the same test for the 2 or 3 competitors who appear in AI responses instead of you. Note which platforms cite them, which query types trigger their recommendations, and what language AI uses to describe them. This competitive snapshot tells you the gap you need to close and on which platforms closing it will move the needle fastest.
What to Look For in Competitor AI Responses
When AI recommends your competitor, it is giving you a roadmap. The language AI uses to describe them tells you which signals it is responding to: "has strong reviews," "recently updated," "licensed and insured in [city]," "specializes in [service type]." Each descriptor points to a signal gap your competitor has closed that you have not. That gap is your optimization priority.
One pattern worth tracking: whether AI is recommending national chains or aggregator platforms (like Yelp category pages) instead of specific local businesses. When AI cites a category page instead of a named business, that signals the entire local market has weak entity signals. Getting your business entity properly recognized in that context can move you into the recommendation slot faster than in markets where established local businesses already dominate AI citations.
Competitor AI Benchmark Scorecard
| Competitor | Platforms Cited On | Language AI Uses | Your Gap |
|---|---|---|---|
| Competitor A | Record after testing | Describe AI tone here | Note what they have that you do not |
| Competitor B | Record after testing | Describe AI tone here | Note what they have that you do not |
What to Do After You Run the Test
If you scored below 40: the priority is entity clarity and review profile. Getting your foundation right takes precedence over optimization tactics. If you scored 40-80: identify your two weakest platforms and examine what signals they weight differently. If you scored above 80: you have a visibility foundation worth protecting. Focus on expanding the query types you appear in and monitoring competitor positioning monthly.
Find Out What AI Says About Your Business Right Now
Our Blind Spot Report does the 4-platform test for you, interprets the results, and tells you exactly which visibility gaps are costing you citations. No manual testing required.
Get Your Free Blind Spot ReportFrequently Asked Questions
What is the fastest way to check if ChatGPT knows my business?
Open ChatGPT and type: "Who is the best [your business type] in [your city]?" or "Can you recommend a [your service] near [your location]?" If your business name does not appear in the first response, that is a strong signal you are not being cited. Run the same test on Perplexity and Google AI Overviews, as each platform has different citation logic and you may appear on one but not others.
Why would I show up on Perplexity but not ChatGPT?
Only 11% of domains cited by ChatGPT are also cited by Perplexity, meaning the two platforms have almost entirely different citation logic. Perplexity heavily weights freshness and live web search results. ChatGPT weights entity recognition, review signals, and its training data. A business with recent content and reviews might rank on Perplexity before it appears on ChatGPT.
What does it mean if AI mentions my business but does not recommend it?
There is a meaningful difference between "mentioned" and "recommended." Being mentioned means AI is aware your business exists. Being recommended means AI is confident enough in your trustworthiness and relevance to cite you when someone asks for help. Businesses stuck at "mentioned" typically have partial visibility signals but have not closed the trust and specificity gap that triggers actual recommendations.
How often should I run an AI visibility check?
Monthly is a practical baseline for most small businesses. Weekly checks make sense during an active optimization push when you want to track whether changes are producing early signals. AI platforms update their indexes continuously, so a snapshot from three months ago may no longer reflect your current standing.
What should I do if I am invisible on all four AI platforms?
Complete invisibility across all platforms is a signal of a foundational gap: your business entity is not clearly defined or consistently recognized across the sources AI reads. The most effective first step is a structured audit that maps your current entity clarity, review profile, schema coverage, and citation footprint, then identifies which gaps are causing the invisibility.
Can I improve my AI visibility without changing my website?
Partially. You can improve your review profile, directory presence, and Google Business Profile without touching your website. These are meaningful signals. However, website-level changes like adding structured schema markup, improving content extractability, and creating FAQ pages with question-and-answer schema tend to produce the most durable citation improvements.
Know Exactly Where You Stand on AI Search
The 4-platform test tells you your current visibility score. The Blind Spot Report tells you why your score is what it is and what to fix first. Get yours free.
Get My Free Blind Spot ReportFree analysis. No credit card required. Results in 24 hours.
AEO for Real Estate Agents
The playbook we built on our own real estate site — 1.14M+ monthly impressions, cited by 4 of 4 AI platforms. Now available to one agent per market.
See the real estate playbook →Free // 30-Minute AuditRun a Free Answer Engine Audit
See how ChatGPT, Claude, Perplexity, and Google AI answer questions about your category today — and where your name should be.
Book the audit →