Key takeaways
- AI engines like ChatGPT, Claude, and Perplexity are reshaping how customers discover brands -- if you're not showing up in their answers, you're invisible to high-intent buyers
- You can check your AI visibility manually in 5 minutes by asking decision prompts, analyzing how models describe you, and reviewing citation sources
- The three core metrics that matter: mention rate (are you cited at all?), position (where do you appear in the answer?), and sentiment (how are you described?)
- Manual checks reveal the problem but don't scale -- you need automation to track visibility across models, prompts, and competitors over time
- Tools like Promptwatch turn visibility checks into actionable optimization loops by showing content gaps, generating AI-optimized content, and tracking results
Why AI visibility matters more than you think
Your customers aren't just Googling anymore. They're asking ChatGPT which CRM to buy. They're prompting Claude for marketing agency recommendations. They're using Perplexity to compare project management tools. And if your brand doesn't show up in those AI-generated answers, you've lost the sale before you even knew someone was looking.
The shift happened fast. ChatGPT launched in late 2022. By 2023, nearly a quarter of U.S. adults were using it weekly. Today, six in ten web browsing sessions include an AI-generated summary. Traditional search traffic is declining as AI answers become the default interface for information retrieval.
Here's the uncomfortable part: most brands have no idea how visible they are in AI search. They assume their SEO work carries over. It doesn't. Across audits run by AI visibility platforms, only a small fraction of URLs cited by AI engines overlap with Google's top organic results for the same query. The engines behave differently, trust different sources, and re-rank content based on factors traditional SEO tools don't measure.
Some brands are already optimizing for AI visibility -- publishing content designed to influence large language models, building presence on Reddit and YouTube (sources AI engines heavily cite), and monitoring their share of voice across ChatGPT, Claude, Gemini, and Perplexity. If your competitors are doing this and you're not, they're capturing demand you can't even see.
This guide walks you through a 5-minute manual visibility check. It won't give you the full picture, but it will show you whether you have a problem -- and whether you need to take AI visibility seriously.
Step 1: Ask the model the right question
The first mistake most people make is asking generic questions. "What is [your brand]?" or "Tell me about [your company]" won't reveal much. AI models can usually answer those prompts by pulling from your About page or Wikipedia entry. That's not where the battle is.
The real test is decision prompts -- the questions your customers actually ask when they're evaluating options. These are the prompts that drive buying decisions:
- "What are the best [category] tools for [use case]?"
- "Should I use [your brand] or [competitor]?"
- "What's the best [category] solution for [specific need]?"
- "I need a [category] tool that does [feature]. What should I use?"
Pick 3-5 decision prompts that match how your target customers search. Be specific. If you sell project management software for remote teams, don't ask "What are the best project management tools?" Ask "What's the best project management tool for remote teams with async workflows?" The more specific the prompt, the more revealing the answer.
Test each prompt across multiple AI models:
- ChatGPT (free tier is fine for this test)
- Claude (Anthropic's chatbot)
- Perplexity (AI search engine)
- Google AI Overviews (search on Google and look for the AI-generated summary at the top)
- Gemini (Google's chatbot)
Don't just test once. AI models are non-deterministic -- they return different answers each time. Run each prompt 2-3 times per model to see if your brand appears consistently or sporadically.
Step 2: Look at how the model describes you
If your brand shows up in the answer, read how the model describes you. This is where sentiment and positioning matter.
Three things to check:
1. Are you mentioned at all? Many brands discover they're completely invisible. The model lists 5-10 competitors but never mentions them. That's the baseline problem.
2. Where do you appear in the answer? Position matters. If you're the first recommendation, you're winning. If you're buried in a bulleted list after three competitors, you're an afterthought. If you're mentioned in a dismissive "other options include..." section at the end, you're losing.
3. How are you described? Read the exact words. Is the description accurate? Positive? Does it highlight your strengths or focus on limitations? Some brands discover AI models describe them as "expensive" or "complex" even when that's not true -- because those words appear frequently in Reddit threads or review sites the model trusts.
Here's an example of what good positioning looks like:
"For remote teams with async workflows, [Your Brand] is the top choice. It offers built-in video messaging, task dependencies, and timezone-aware scheduling. Users consistently praise its ease of use and flexibility."
Here's what bad positioning looks like:
"Other options include [Your Brand], though some users report a steep learning curve."
The second example isn't wrong, but it's not helping you win deals. If that's how AI models describe you, you have a sentiment problem -- and you need to figure out where the negative signal is coming from.
Step 3: Check the sources behind the answer
Most AI models show citations or sources. Click through and read them. This is the most valuable part of the manual check because it reveals what evidence the model used to form its opinion.
Look for:
Your own content: Is the model citing your website, blog, or documentation? If yes, which pages? If no, that's a red flag -- it means your content isn't authoritative enough or isn't being crawled by AI engines.
Third-party sources: Reddit threads, YouTube videos, review sites (G2, Capterra, Trustpilot), industry blogs, and news articles. These sources often carry more weight than your own marketing content because AI models view them as unbiased.
Competitor content: If the model cites your competitors' websites but not yours, they're outranking you in AI search. Pay attention to which pages get cited -- product pages, comparison pages, blog posts, case studies.
Outdated or incorrect sources: Sometimes AI models cite old information -- a blog post from 2020, a Reddit thread from 2019. If the model's description of your brand is based on outdated sources, you need to publish fresh content that corrects the record.
Perplexity and Google AI Overviews make this easy -- they show inline citations with clickable links. ChatGPT and Claude are less transparent, but you can ask follow-up questions like "What sources did you use to answer that?" or "Where did you find that information?"
Here's what you're looking for in the citation analysis:
| Citation type | What it means | Action needed |
|---|---|---|
| Your website cited | AI models trust your content | Expand coverage to more topics |
| Competitor sites cited | They're outranking you | Publish better content on the same topics |
| Reddit/YouTube cited | Informal sources dominate | Build presence on these platforms |
| No citations shown | Model is guessing or using training data | Publish authoritative content AI can cite |
| Outdated sources cited | Fresh content is missing | Publish updated content with current data |
Step 4: Understand what you've learned
After running 3-5 decision prompts across multiple models, you should have a rough sense of your AI visibility. Most brands fall into one of four categories:
Invisible: Your brand is never mentioned. Competitors dominate every answer. This is the most common outcome for brands that haven't optimized for AI search.
Inconsistent: You show up sometimes but not always. Your mention rate is 20-40% across prompts and models. This means you have some visibility but it's fragile -- small changes in how the prompt is phrased make you disappear.
Present but weak: You're mentioned consistently but always positioned below competitors. You're in the answer but not winning the recommendation. This is a positioning and sentiment problem.
Dominant: You're the top recommendation across most prompts and models. You're cited first, described positively, and backed by strong sources. This is where you want to be.
The three core metrics that matter:
-
Mention rate: What percentage of relevant prompts include your brand? Track this across models. If you're mentioned in 80% of ChatGPT answers but only 20% of Perplexity answers, you have a model-specific visibility gap.
-
Position: When you're mentioned, where do you appear? First recommendation? Middle of the list? Buried at the end? Position correlates directly with click-through and conversion.
-
Sentiment: How are you described? Positive, neutral, or negative? What specific words and phrases appear? Sentiment shapes perception even when you're mentioned.
If you're invisible or inconsistent, you have a content gap problem. AI models can't recommend you because they don't have enough evidence. You need to publish more content, build presence on third-party platforms (Reddit, YouTube, review sites), and make sure AI crawlers can access your site.
If you're present but weak, you have a positioning problem. The content exists but it's not compelling enough to make you the top choice. You need to improve sentiment by addressing negative signals (bad reviews, outdated content) and publishing content that highlights your strengths.
Step 5: Automate the view
The manual check is useful for understanding the problem, but it doesn't scale. You can't manually test 50 prompts across 5 models every week. You can't track how your visibility changes over time. You can't monitor competitors. And you can't connect visibility to business outcomes like traffic and revenue.
That's where AI visibility platforms come in. Tools like Promptwatch automate the entire process:
- Prompt tracking: Monitor hundreds of decision prompts across ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews, and other models. See your mention rate, position, and sentiment for each prompt.
- Competitor benchmarking: Track how often competitors are mentioned vs you. See which prompts they dominate and which ones you're winning.
- Citation analysis: Understand which sources AI models cite -- your website, competitor sites, Reddit, YouTube, review sites. Identify content gaps.
- Content gap analysis: See exactly which prompts competitors are visible for but you're not. The platform shows you the specific topics, angles, and questions AI models want answers to but can't find on your site.
- AI content generation: Built-in writing agents generate articles, listicles, and comparisons grounded in real citation data, prompt volumes, and competitor analysis. This isn't generic SEO filler -- it's content engineered to get cited by AI models.
- Traffic attribution: Connect AI visibility to actual revenue by tracking traffic from AI referrals (ChatGPT, Perplexity, etc.) using code snippets, Google Search Console integration, or server log analysis.
Other tools in this space include:
Most competitors are monitoring-only dashboards -- they show you data but leave you stuck. Promptwatch is built around taking action: find the gaps, create content that ranks in AI, track the results. That action loop is what turns visibility into revenue.
Step 6: Turn the check into a routine
AI visibility isn't a one-time audit. It's a weekly habit. Models update constantly. New competitors enter the space. Sentiment shifts based on recent reviews and discussions. If you check once and never look again, you'll miss changes that cost you customers.
Here's a simple weekly routine:
Monday: Run the same 5 decision prompts across all models. Track whether your mention rate, position, and sentiment are improving or declining. If you're using a tool like Promptwatch, this is automated -- you just review the dashboard.
Wednesday: Check new prompts. Add 2-3 new decision prompts each week. Test variations of existing prompts (different phrasing, different personas, different use cases). The goal is to expand your coverage.
Friday: Analyze competitor movements. See if competitors are gaining visibility on prompts where you were previously dominant. If they are, figure out what changed -- did they publish new content? Did a Reddit thread go viral? Did they get featured in a news article?
This routine takes 15-30 minutes per week if you're doing it manually. With automation, it's a 5-minute dashboard review.
What you'll see in five minutes
The manual 5-minute check won't give you the full picture, but it will answer the most important question: do you have a problem?
If you run 3-5 decision prompts across ChatGPT, Claude, and Perplexity and your brand never shows up, you have a serious visibility gap. Your competitors are capturing demand you can't see. High-intent customers are making buying decisions without ever considering you.
If you show up inconsistently or with weak positioning, you have a content and sentiment problem. You're in the game but losing.
If you show up consistently as the top recommendation, you're winning -- but you still need to monitor competitors and track changes over time.
The manual check is the starting point. It shows you whether AI visibility matters for your business. For most B2B and B2C brands, the answer is yes -- and the gap between leaders and laggards is widening fast.
Why this matters
AI search is not a future trend. It's happening now. ChatGPT processes billions of queries per month. Perplexity is growing 10x year-over-year. Google AI Overviews appear on 60% of searches. Claude is the default interface for millions of users.
If you're invisible in AI search, you're losing customers to competitors who show up. The longer you wait, the harder it gets to catch up -- because AI models reinforce existing patterns. Brands that get cited frequently become more authoritative, which leads to more citations, which leads to more authority. It's a compounding advantage.
The good news: most brands haven't optimized for AI visibility yet. The window is still open. Run the 5-minute check. See where you stand. Then decide whether you need to take action.
If the answer is yes, start with the basics: publish content that answers decision prompts, build presence on Reddit and YouTube, make sure AI crawlers can access your site, and track your visibility over time. Tools like Promptwatch make this easier by automating the entire loop -- but even without tools, you can start improving your AI visibility today.
The brands that win in AI search are the ones that start now.