Not sure where to start? Take a guided tour.
AI Site Grader
AI Query Scanner
llms.txt GeneratorThe short answer
This issue occurs when your brand data is heavily represented in Common Crawl or Reddit (which OpenAI favors) but lacks the real-time search signals or technical formatting required by other models. The solution involves diversifying your backlink profile and implementing platform-specific schema.
- Quickest fix
- Submit your sitemap directly to Bing Webmaster Tools and Google Search Console to refresh the live-search indexes used by Perplexity and Gemini.
- Most common cause
- Over-reliance on historical training data that ChatGPT possesses, while lacking the real-time search authority needed for RAG-based models like Perplexity and Gemini.
Diagnosis
Symptoms
- Brand appears in ChatGPT responses but is missing from Perplexity citations
- Claude claims to have no information about your company
- Gemini prioritizes competitors even when your product is a better fit
- Search-enabled AI models fail to browse your site successfully
How to confirm it
- Run the same prompt across ChatGPT, Perplexity, Claude, and Gemini
- Check if your site allows the ‘User-agent: *’ or specific AI bots like ‘Google-Extended’ and ‘ClaudeBot’
- Verify if your brand mentions are limited to older forum threads (pre-2023)
- Severity
- Medium
- Urgency
- High if you rely on top-of-funnel discovery
- Impact
- Loss of market share as users migrate from ChatGPT to more research-oriented tools like Perplexity.
Common causes
| Cause | How to identify it | How common | Effort |
|---|---|---|---|
| Training data vs. RAG disconnect | ChatGPT knows you but Perplexity (using live search) does not. | Very common | Medium fix |
| Aggressive robots.txt blocking | Check your robots.txt for ‘Disallow: /‘ on ClaudeBot or OAI-SearchBot. | Common | Easy fix |
| Low domain authority in bing/Google | Check if your site ranks on page 3+ for your own brand name in standard search. | Common | Hard fix |
| Lack of structured data (JSON-LD) | Run your URL through the Schema.org validator; check for Organization and Product types. | Sometimes | Medium fix |
| Format incompatibility | Your content is locked behind heavy Javascript or PDFs that certain bots struggle to parse. | Rare | Medium fix |
Quick wins
Update your Crunchbase and LinkedIn profiles with detailed descriptions.
Models like Claude and Gemini often use these as ‘seed’ data for brand identity.. Time: 30 minutes
Submit your URL to Perplexity’s ‘Suggest a Source’ if applicable or ensure it is indexed in Bing.
Improved visibility in Perplexity’s ‘Sources’ section.. Time: 10 minutes
Remove ‘Disallow: /‘ for Google-Extended in robots.txt.
Enables Gemini to use your site content for its responses.. Time: 5 minutes
Solutions
Optimize for live search API (RAG)
Timeline: 1–2 weeks
Improve Bing Indexing
Since Perplexity and others use Bing/Google APIs, use Bing IndexNow to push updates instantly.
Tools: Bing Webmaster Tools, IndexNow API
Boost Real-time Citations
Get mentioned in recent (last 30 days) news articles or high-authority industry blogs.
Tools: PR tools, Guest Posting
Audit and update bot permissions
Timeline: 24–48 hours
Identify Missing Bots
Specifically allow ClaudeBot, Google-Extended, and PerplexityBot in your robots.txt file.
Tools: Text Editor, FTP
Verify Access
Use a bot simulator to ensure these crawlers aren’t being blocked by your CDN/WAF.
Tools: Cloudflare Dashboard, Log analysis
Implement LLM-friendly schema
Timeline: 1 week
Deploy Organization Schema
Add detailed JSON-LD including ‘sameAs’ links to social profiles and Wikipedia.
Tools: Schema Pro, Google Rich Results Test
Add Product/Service Markup
Ensure prices, features, and reviews are clearly tagged for Gemini’s shopping/comparison engine.
Tools: JSON-LD Generator
Diversify source citation profile
Timeline: 2–4 weeks
Target Alternative Databases
Get listed on G2, Capterra, and Trustpilot, as Gemini and Claude frequently pull from these ‘truth’ sources.
Tools: Review Management Software
Update Crunchbase/LinkedIn
Many models use these as foundational identity layers.
Tools: Crunchbase, LinkedIn Company Page
Conversion to semantic content structure
Timeline: 1 week
Flatten Site Architecture
Ensure key brand info is in HTML text, not buried in images or complex JS components.
Tools: Screaming Frog
Create an ‘AI Fact Sheet’
Publish a /press or /about page with clear, bulleted facts that bots can easily scrape.
Tools: CMS
Localized SEO for gemini/Google AI
Timeline: 2 weeks
Optimize Google Business Profile
Gemini heavily weights Google Maps and Business data for local/service queries.
Tools: Google Business Profile
Internal Linking Refresh
Use descriptive anchor text for internal links to help Google’s AI understand page hierarchy.
Tools: Search Console
Example scenarios
These scenarios show how the fix plays out. They are illustrative, not customer results.
- Situation
- A SaaS startup was the top recommendation on ChatGPT but was completely absent from Perplexity.
- Diagnosis
- The brand had high historical mentions on Reddit (training data) but zero presence in Bing’s top 100 search results (RAG data).
- Fix
- Focused on a 30-day PR sprint to get mentioned in 5 tech publications indexed by Bing.
- Result
- Perplexity began citing the brand in 40% of relevant queries.
- Lesson
- ChatGPT’s memory is long, but Perplexity’s memory is based on current search authority.
- Situation
- An e-commerce brand appeared in ChatGPT but Gemini refused to show their products.
- Diagnosis
- The site was blocking ‘Google-Extended’ and lacked Product Schema.
- Fix
- Unblocked the bot and added detailed JSON-LD for the product catalog.
- Result
- Gemini started including product cards with prices in its responses.
- Lesson
- Google’s AI models require specific technical permissions and structured data.
- Situation
- A B2B consultancy was visible in ChatGPT but Claude claimed it ‘didn’t exist’.
- Diagnosis
- Claude relies heavily on high-trust web directories which the company had ignored.
- Fix
- Claimed and optimized profiles on G2, LinkedIn, and official industry registries.
- Result
- Claude was able to summarize the company’s services accurately.
- Lesson
- Anthropic’s models favor curated, high-trust databases over general web noise.
Prevention
Strategies
- Maintain a ‘Bot-Friendly’ robots.txt policy by default.
- Distribute PR and content updates across platforms, not just one niche.
- Regularly audit your brand’s presence in Bing and Google search results.
Monitoring setup
- Weekly automated prompts across the ‘Big 4’ AI platforms.
- Track ‘AI Referrals’ in your analytics traffic sources.
- Monitor Bing Webmaster Tools for crawl errors.
Warning signals
- A drop in ‘Direct’ traffic that correlates with AI search usage spikes.
- Competitors appearing in Perplexity citations while you are omitted.
- Inaccurate brand facts appearing in Gemini but not ChatGPT.
When to escalate
- Signs
- You have 90+ Domain Authority but still aren’t cited.
- Your site is fully indexed but AI models explicitly state they are ‘blocked’ from viewing it.
- Competitors are getting ‘rich cards’ or special formatting you can’t access.
- Options
- Consult an AIO (AI Optimization) specialist.
- Engage a technical SEO agency to audit crawlability.
- Reach out to platform developer forums (e.g., OpenAI/Google Cloud).
- Timeline
- If no change is seen after 6 weeks of consistent optimization.
Frequently asked questions
ChatGPT relies heavily on a static training dataset that might include your brand from older web crawls (like Common Crawl). Perplexity uses Retrieval-Augmented Generation (RAG), which means it searches the live web (primarily via Bing) every time you ask a question. If your site isn’t ranking well in live search or has poor technical SEO, Perplexity won’t find you even if ChatGPT knows you by heart.