How to Check If Your Business Appears in AI Search

Most businesses have no idea what AI says about them. Here's how to find out.

Your customers are asking AI engines questions about your industry, your competitors, and your solutions. ChatGPT has 800 million weekly users. Perplexity has reached 45 million active users. Google's AI Overviews now appear in 1.5 billion monthly searches. The question isn't whether AI is influencing buying decisions anymore, it's whether your business is visible when it matters.

This guide walks you through the practical steps to check your AI search visibility across every major platform, understand what results you're getting, and identify gaps in your visibility.

How do you check if your business appears in AI search?

The most straightforward way to check your AI search visibility is to ask the AI engines directly, using the same prompts your customers would use. There's no dashboard, no API, and no official tool. You test manually by typing prompts into each platform and recording what the AI says about your business, your competitors, and your category.

The manual testing process has five basic steps:

Step 1: Identify your core search scenarios. These are the prompts customers actually search for. For a healthcare software company, this might be "best patient management systems for small clinics". For an accounting firm, it's "affordable tax preparation services in Melbourne". For a SaaS product, it's "project management tools similar to Asana". Write down 10 to 15 realistic prompts that represent how customers would find you.

Step 2: Test across platforms. Each AI engine has different training data, citation patterns, and retrieval mechanisms. ChatGPT, Perplexity, Google AI Overviews, and Gemini will give you different results. Testing one platform tells you nothing about your actual AI visibility.

Step 3: Record the results. Note whether your business is mentioned, what position it appears in, whether it's cited with a source, what sentiment is expressed, and which competitors appear instead. A spreadsheet or document works fine for this.

Step 4: Check the citations. When an AI engine cites your business, where did that information come from? This tells you whether the AI has access to your brand-managed content (your website, case studies, press releases) or whether it's pulling from third-party sources (review sites, directories, competitors' websites).

Step 5: Look for gaps and errors. Is your business completely missing from results where competitors appear? Are there factual errors about your services, pricing, or location? Are competitors being cited more frequently? These gaps are where you can improve your AI visibility.

This manual approach gives you a baseline, but it's time-intensive and doesn't scale. You'll test a handful of prompts today and miss dozens of variations that customers actually search for. You'll test once and won't know if your visibility improved or declined over time. That's why we'll also cover when to escalate to professional auditing later in this guide.

What prompts should you test?

The prompts you test are the foundation of meaningful results. If you only test one generic prompt, you'll miss visibility opportunities across dozens of realistic customer search scenarios.

Start by mapping the customer journey. When someone is considering your solution, what question do they ask first? What do they ask before they compare? What do they ask before they make a final decision? Each of these questions represents a different search prompt.

For a productivity software company, the journey might look like this:

• Awareness: "best project management tools 2026"
• Consideration: "Asana vs Monday.com vs ClickUp"
• Decision: "project management software for remote teams with budget tracking"

For a financial advisory firm, it might be:

• Awareness: "independent financial advisors in London"
• Consideration: "financial advisors who specialise in tax-efficient investing"
• Decision: "fee-only financial planners for FTSE executives"

Each stage needs testing. The AI engines will return completely different results, and your business might be visible in one journey stage and invisible in another.

Industry-specific examples help clarify this:

SaaS and software: Test "best [category] for [use case]", comparison prompts ("X vs Y"), and integration-specific prompts ("best CRM for HubSpot integration"). Test problem-first prompts too ("how to automate invoice processing").

Professional services: Test location-plus-service prompts ("accountants in Manchester specialising in startup tax"), credential-specific prompts ("qualified financial planners near Bristol"), and niche-specific prompts ("solicitors specialising in family law").

E-commerce and retail: Test product category prompts ("best wireless headphones for gym"), buying intent prompts ("where to buy [your product]"), and comparison prompts ("alternatives to [competitor]").

Healthcare: Test symptom-based prompts ("doctors near me for lower back pain"), service-specific prompts ("best clinics for sports injury near Dublin"), and credential-specific prompts ("physiotherapists specialising in ACL rehab").

Your initial list should include 10 to 15 prompts. Include your brand name in some, exclude it in others. Include location if you're location-dependent. Include comparison prompts if you compete head-to-head. Include problem-first prompts if customers search by problem rather than solution.

How to test your business on ChatGPT

ChatGPT is the entry point for most AI visibility testing. It has 800 million weekly active users and influences purchasing decisions across nearly every industry.

To test on ChatGPT, you'll need a free or paid account at chat.openai.com.

Step 1: Open ChatGPT and select your model. At the bottom of the page, you can choose between available models. For the most realistic testing, use GPT-4o, which is more recent and more widely used than earlier models. Free users may have access to GPT-4 Mini depending on account age.

Step 2: Set the conversation context (optional but recommended). Before you ask your first prompt, you can optionally give ChatGPT context about your goals. For example: "I'm testing AI search visibility for a B2B SaaS company. I'll be asking you several prompts and want you to recommend the best tools based on features and use cases, not just listing options." This gives more realistic results than generic prompts.

Step 3: Enter your test prompt. Type one of your core search prompts exactly as a customer would type it. If your prompt is "best project management tools for small teams", type exactly that. Don't add context about it being a test. ChatGPT will return a list of recommendations, usually in order of prominence.

Step 4: Record the results. Take a screenshot. Note:

• Whether your business is mentioned in the initial response
• What position it appears in (first mention, second, third?)
• Whether it includes a web citation (you can hover over links to see the source)
• What description ChatGPT gives of your business
• What competitors appear and in what order

Step 5: Ask follow-up prompts. ChatGPT's follow-up responses often differ from initial responses. Try follow-ups like:

• "Tell me more about [your competitor]"
• "What makes [your company name] different from [competitor]?"
• "Which of these integrates best with [related tool]?"
• "What do customers say about [your company name]?"

These follow-ups reveal whether ChatGPT has nuanced knowledge of your business or just surface-level data.

Step 6: Test comparison prompts. Enter prompts like "[Your company] vs [Competitor]" or "[Your company] vs [Competitor] vs [Competitor]". Note whether ChatGPT knows the specific differences and whether it positions your company favourably or unfavourably.

ChatGPT visibility red flags:

• Your business is never mentioned across any of your test prompts
• Your business is only mentioned when directly named in the prompt
• ChatGPT gets basic facts wrong (pricing, location, features)
• Competitors are consistently mentioned first and more frequently
• ChatGPT can't answer questions about your specific capabilities
• Sources are pointing to outdated content or competitor pages

How to test your business on Perplexity

Perplexity is increasingly important for AI visibility testing. It has 45 million active users and emphasises citation transparency more than ChatGPT does. Every claim includes a visible source link.

To test on Perplexity, go to perplexity.ai. You can use the free version for testing.

Step 1: Select your research mode. Perplexity offers different modes including Copilot (most thorough), Writing (for content generation), and Collections (for saved research). For visibility testing, use the default Copilot mode, which retrieves the most sources.

Step 2: Enter your prompt. Type the same test prompts you used for ChatGPT. Perplexity will instantly return results with numbered source citations visible on the right side of the screen.

Step 3: Pay attention to the citation sources. This is where Perplexity differs significantly from ChatGPT. Every claim has a visible source number. Click through to see where Perplexity sourced the information. Are these:

• Your official website?
• Brand-controlled content (case studies, white papers)?
• Third-party review sites?
• Competitor websites?
• News coverage?

The source distribution tells you whether you control the narrative about your business in AI search.

Step 4: Check the source count. Count how many sources mention your business versus competitors. Perplexity might cite your competitor 7 times and your business once. This imbalance is more visible here than in ChatGPT.

Step 5: Look for the "Sources" tab. At the top right, Perplexity shows a "Sources" summary. This widget reveals how many sources it consulted overall and which domains appear most frequently. If Competitor.com appears 5 times and YourBusiness.com appears zero times, that's a clear visibility gap.

Step 6: Test follow-up searches. Search variations like:

• "[Your company] reviews"
• "[Your company] pricing"
• "[Your company] customers"
• "[Your company] case studies"

Perplexity should surface your official content for these searches if you have good AI visibility.

Perplexity visibility red flags:

• Your official website never appears in the sources
• Competitors' domains appear far more frequently than yours
• Most citations come from review aggregators, not your own content
• Perplexity can't find pricing or features information about you
• Results mention your business but cite competitor sources for the information
• Your business has zero sources but competitors have multiple

How to test your business on Google AI Overviews

Google AI Overviews appear in 1.5 billion monthly searches and cause a 61% drop in organic click-through rates to websites below the overview. If your business is cited in an overview, you earn 35% more organic clicks than businesses not mentioned.

Testing Google AI Overviews requires a Google Search account. Go to google.com and perform a search.

Step 1: Search for your test prompt. Enter one of your core prompts into Google's search bar. For example, "best project management tools for small teams".

Step 2: Look for the AI Overview section. At the top of the search results, above the traditional organic results, you'll see a new section with a light purple/blue background. This is the AI Overview. Not every search triggers an AI Overview, some searches still return traditional organic results only. If no overview appears, try variations of the prompt. Sometimes "best X for Y" prompts return overviews while "how to choose X" prompts don't.

Step 3: Record whether your business is mentioned. The AI Overview typically mentions 3-5 businesses. Note:

• Is your business mentioned at all?
• What position does it appear in?
• Is there a link to your website?
• What specific claims does the overview make about you?

Step 4: Click through to the full overview. Below the initial overview excerpt, there's a "Read more" or "Show more" option. Click this to see the full AI Overview, which may mention additional businesses and provide deeper context.

Step 5: Check the sources. Google AI Overviews include source links inline and in a sidebar. Click these to verify:

• Are they linking to your official website?
• Are they linking to competitor websites, review sites, or news coverage?
• Is the information current?

Step 6: Test location-based overviews. Search for location-specific prompts like "best restaurants near me" or "accountants near Colchester". Location-based AI Overviews often have different business mentions than national searches.

Google AI Overview visibility red flags:

• Your business never appears in any AI Overview
• Competitors appear in overviews but you don't
• The overview cites information about you from competitor sources
• Your website isn't linked even when you're mentioned by name
• The location-based overviews exclude your business despite location relevance
• Information in the overview is outdated or inaccurate

How to test your business on Gemini and other platforms

Beyond ChatGPT and Perplexity, test on Google Gemini, Claude, and any platform relevant to your target audience.

Google Gemini:

Gemini is Google's response to ChatGPT and it has different training data and citation patterns. Access it at gemini.google.com (requires a Google account).

The testing approach is identical to ChatGPT, but note that Gemini is more recent and may have more current training data. Gemini also emphasises "Conversing" as a mode, meaning it's designed for back-and-forth dialogue. Ask your initial prompt, then test follow-ups more extensively than you would on ChatGPT.

Claude:

Anthropic's Claude is increasingly used for business research and analysis. Access it at claude.ai (requires an account).

Claude emphasises detailed, nuanced responses over quick lists. Instead of returning a bulleted list of tools, Claude might return a paragraph analysing three tools in depth. Test the same prompts on Claude, but pay attention to how it frames your business. Does it give nuanced context about your advantages? Does it acknowledge limitations competitors have that you don't?

Specialised platforms:

If your business operates in a specific vertical, test on platform-specific AI:

• Reddit's integrated AI search function
• LinkedIn's search (which now includes AI-powered recommendations)
• Amazon's AI search for e-commerce businesses
• TripAdvisor's AI search for hospitality
• Any vertical-specific AI platform your customers use

The process remains the same: type realistic prompts, record whether you're mentioned, check source citations, and look for gaps.

What should you look for in the results?

Raw visibility data (whether you're mentioned or not) is only the starting point. Understanding how you're positioned, what information is accurate, and how you compare to competitors reveals the true state of your AI search visibility.

Citations and sources matter most. When an AI engine mentions your business, it typically includes a source link. Check:

• Is the source your official website, or a third party?
• Is the source current (updated in the past 6 months)?
• Does the source accurately represent your current business?

If 80% of your citations are from a LinkedIn profile you haven't updated since 2023, that's a problem. If all your citations come from review aggregators, you're not controlling your narrative. Ideally, 60-70% of your citations should come from brand-managed sources: your website, official press releases, or recent news coverage you've influenced.

Accuracy is a visibility metric. When an AI engine describes your business, are the facts correct? Check:

• Pricing: Is it current or outdated?
• Location: Is your primary location listed correctly?
• Services: Does it describe all your main services or just one?
• Company size: Does it accurately represent whether you're a startup or enterprise?
• Differentiators: Does it mention what makes you unique?

Inaccurate information in AI search is a visibility problem that reduces click-through rates and conversion. Customers who see outdated pricing on ChatGPT won't click through to your website.

Sentiment and framing reveal positioning. Beyond accuracy, how does the AI frame your business compared to competitors?

Read the descriptions carefully. Does the AI say:

• "X is a good choice for [use case]" (positive, specific)
• "X is one of the options" (neutral, non-specific)
• "X is less popular than Y but offers [advantage]" (comparative, honest)
• "X is outdated compared to newer tools" (negative, problematic)

An AI engine might mention your business but frame it as a legacy option, which damages visibility even though you're technically "mentioned".

Competitor visibility gaps show opportunities. Note which competitors appear in your test results and which don't. If Competitor A appears in every AI overview but Competitor B (who you consider equally important) appears in none, that's a visibility lesson.

Why does Competitor A appear more often? Often because:

• They have more brand-managed content (website, case studies, white papers)
• They have more recent online presence and recent content
• They have more third-party citations and mentions
• They explicitly optimise for AI discoverability

These gaps are addressable. They're not random; they reflect content strategy and AEO optimisation.

Breadth of mention is important. Being mentioned in one AI engine while completely invisible in three others is a critical gap. Track:

• ChatGPT: Visible or not visible?
• Perplexity: Visible or not visible?
• Google AI Overviews: Visible or not visible?
• Gemini: Visible or not visible?
• Specialised platforms: Visible or not visible?

Ideally, you want at least 80% coverage: visible in at least 4 of 5 platforms. If you're only visible on ChatGPT, your AI visibility is dependent on a single source.

Manual testing vs professional auditing

Manual testing serves one critical purpose: it gives you immediate baseline visibility. You can test your business on ChatGPT right now, in 15 minutes, and know whether customers find you there. That's valuable.

But manual testing has three major limitations:

One. You test a handful of prompts when thousands of variations exist. You test "best project management tools" once and never test variations like "project management tools for freelancers" or "cheapest project management software 2026" or "project management tools with Slack integration". You're testing 0.5% of the search scenarios that matter. For a more comprehensive approach, consider learning about multi-platform optimization.

Two. You test once and can't track trends. Did your visibility improve last month? You'd have to manually repeat the entire test to know. Most businesses never retest, meaning they don't know whether their visibility improved or declined after they've made changes.

Three. You're looking at results but not understanding why they happen. The AI mentioned Competitor A three times and you once. But why? Was it because Competitor A has fresher content? More citations from authority sources? Better brand mentions in news coverage? Manual testing shows the symptom but not the diagnosis.

This is where professional AI visibility audits become critical. An audit tool tests hundreds of prompt variations across multiple platforms, tracks visibility month-over-month, and identifies the specific content and citation factors driving your visibility.

At Omni Eclipse, we use the Eclipse Score for this purpose. The Eclipse Score is a proprietary methodology that measures your visibility across AI search engines on 500+ prompt variations, compares you to competitors, and identifies the specific content and optimisation gaps you need to address.

You can also use the Eclipse Radar tool, which provides ongoing monitoring of your AI search visibility and automatic alerts when visibility changes, new competitors emerge, or new citation sources appear.

For a complete visibility picture, start with manual testing to establish a baseline, then move to professional auditing for ongoing visibility monitoring and actionable recommendations.

How often should you check your AI search visibility?

If you're testing manually, check your visibility quarterly (every 3 months). This frequency is enough to catch major visibility changes without requiring excessive time investment. However, you should check immediately after:

• Publishing significant new content (website redesign, new case studies)
• Making changes to your SEO structure (title tags, meta descriptions)
• Launching a new product or service
• A major company event or news coverage
• A competitor launch that affects your category

If you're using professional monitoring tools like Eclipse Radar, visibility should be checked continuously. The tool checks your visibility on hundreds of prompts daily and alerts you to changes instantly. This is especially critical because:

1. AI training data changes frequently. ChatGPT, Perplexity, and Google update their training data regularly. Your visibility can improve or decline without any changes on your end.

2. Competitors are optimising too. If competitors improve their AI visibility while you stay static, you gradually lose visibility share even if you haven't changed anything.

3. Prompt variations matter. Your visibility on "best project management tools" might be stable, but your visibility on "cheapest project management software" or "best project management software for agencies" might be declining. You'd never know without testing those specific variations.

4. Speed matters in emerging categories. AI visibility compounds over time. Early optimisation for new product categories can establish dominance before competitors realise the opportunity.

If your industry is moving quickly or you have significant AI visibility competitors, monthly auditing is the minimum. Quarterly manual checking isn't sufficient to stay competitive in AI search.

Frequently asked questions

Q: If my business doesn't appear in AI search, does that mean I have zero AI visibility?

A: Not necessarily. You might not appear in broad category searches (like "best project management tools") but still appear in niche searches (like "best project management tools for nonprofits") or competitive searches (like "Monday.com vs [your product]"). The answer depends on which prompts you're testing. If you don't appear in any test across any platform, then yes, you have zero meaningful AI visibility. But most businesses have scattered visibility: present in some searches and absent in others.

Q: Should I test ChatGPT with my real account or anonymously?

A: Test with your real account. ChatGPT's responses can vary slightly based on account history and conversation context, but for your purposes, consistency matters more than variations. Use the same account consistently so you can compare results over time. If you want one test with completely fresh context, do one test in a new incognito browsing session with an anonymous account, but don't make this your baseline.

Q: My business appears in AI search, but the information is outdated. How do I fix this?

A: AI engines source information from two places: your brand-managed content (your website, press releases, case studies) and third-party sources (news sites, review sites, directories). If a third-party source has outdated information, contact that source to request an update. But more importantly, ensure your official website has fresh, detailed content about your current services, pricing, and case studies. When you update your website, it usually takes 2-6 weeks before that updated content appears in AI search results. The more often you update your content, the fresher your AI visibility will be.

Q: How much of my visibility depends on SEO versus AEO?

A: The two overlap significantly but serve different functions. Traditional SEO optimises for Google's search ranking algorithm. AEO optimises for AI citation and recommendation. The same content often serves both purposes, but their priorities differ. For SEO, you're optimising for keyword density and backlink profile. For AEO, you're optimising for citation quality and brand authority. We recommend focusing on content quality and completeness first (which benefits both SEO and AEO), then addressing AEO-specific optimisations once your baseline visibility is established. Learn more about the difference in our guide to Answer Engine Optimisation.

Q: Which AI platforms should I prioritise testing on?

A: Prioritise based on your customer demographics. If your customers are corporate decision-makers, test ChatGPT and Gemini first (most used in professional settings). If your customers are younger or more tech-forward, include Perplexity and Claude. Test Google AI Overviews regardless of industry, as they reach 1.5 billion monthly users and have proven impact on click-through rates. If you operate in e-commerce, prioritise Amazon's AI search. If you operate in professional services or recruiting, prioritise LinkedIn's AI. Start with the platforms where your target customers actually search.

Q: If I improve my website content, how long until my AI visibility improves?

A: Most AI models have training data that's 3-6 months old. Some, like Perplexity (which pulls from the live web), update faster. If you publish new content today, you might see it cited in Perplexity within 1-2 weeks, but it might take 6+ months to appear consistently in ChatGPT if your content doesn't gain citations from authority sources in the meantime. Speed increases when you publish content that gets cited elsewhere: in news coverage, industry sites, or through earned media. Content that sits on your website alone takes longer to reach AI models compared to content that gets cited across the web.

---