Recorded prompt
“How can a new website get cited by Perplexity?”
- Product
- ChatGPT
- Surface
- ChatGPT web with web search
- Model
- ChatGPT automatic model (model version not exposed)
- Locale
- en-US (English answer; zh-CN interface)
- Observed answer
- Open session ↗
- Observed
- 15 November 2026, 21:24 CST
- Fresh conversation
- Yes
What the answer said
Getting a new website cited by Perplexity involves a combination of technical SEO, high-quality content, topical authority and visibility across the web. The important distinction is that getting your website crawled is not the same as getting it cited. Perplexity needs to be able to access your pages, find them relevant to a user's question, and select them as sources. There is no guaranteed submission process or publicly documented formula for earning citations. Here's a practical strategy for a new website. 1. Make your website accessible to Perplexity Essential first step Perplexity operates a crawler called PerplexityBot, designed to discover and surface websites in its search results. Check the following: Your website is publicly accessible. Important pages return HTTP 200 responses. Your robots.txt file doesn't block PerplexityBot. Your hosting provider, CDN or firewall doesn't block its crawler. Your pages can be read without requiring a login. Your sitemap.xml is available and submitted to search engines. A basic robots.txt configuration could look like: 纯文本 User-agent: PerplexityBot Allow: / Sitemap: https://example.com/sitemap.xml Only use this if you want to allow Perplexity to crawl your public content. Perplexity also publishes its crawler IP ranges, which can help with firewall configuration. The official documentation confirms that Perplexity recommends allowing its crawler in robots.txt and permitting its published IP addresses. Changes can take up to 24 hours to appear in its systems. Perplexity 2. Publish content that directly answers questions This is probably the most important content strategy. Perplexity answers questions, so create pages that answer the questions your potential audience actually asks. For example, instead of publishing: The ultimate guide to our services Why our company is different Welcome to our website Publish content such as: What is [your product category]? How does [your service] work? How much does [your service] cost? What are the benefits of [your solution]? [Product A] vs [Product B]: What's the difference? What should you consider before choosing [your service]? Structure each article like this: Direct answer: Explain the answer in the opening paragraph. Detailed explanation: Expand on the subject. Facts and figures: Include relevant statistics and evidence. Examples: Demonstrate how the information applies. Sources: Link to authoritative references. Conclusion: Summarise the key findings. Make each page useful as a standalone reference. 3. Build topical authority A new website has little established reputation, so it needs to demonstrate expertise through consistently useful content. For example, if your website specialises in digital marketing, create a collection of interconnected resources: Example: Digital marketing topical authority Digital Marketing Central topic / pillar page SEO Technical SEO, keyword research, on-page SEO Content Marketing Content strategy, blogging, content planning AI Search Perplexity, AI visibility, search citations Analytics Performance, measurement, reporting Link supporting articles back to the relevant central topic page. The objective is to establish a recognisable area of expertise rather than publish hundreds of unrelated articles. 4. Get mentioned by other reputable websites Your own website isn't the only place where your brand should appear. Build a genuine presence through: Industry publications. Relevant professional associations. Independent reviews. Expert interviews. Industry research and reports. Reputable blogs and websites. Relevant community discussions. One particularly valuable approach is to publish original research that other websites might reference. For example: An original industry survey. A comparison of industry trends. A useful dataset. A detailed case study. An independently verifiable report. These resources can generate references and links that help people discover your website. 5. Make your content easy for AI systems to understand Use good technical and editorial practices. Element Recommendation Page titles Clear, descriptive titles Headings Logical H1, H2 and H3 structure Schema markup Relevant Article, Organisation, Product or FAQ schema Author information Identify authors and their expertise Publication dates Show publication and update dates Statistics Include sources and dates Internal links Connect related articles Images Use descriptive alt text Mobile performance Ensure pages work well on mobile One important point: structured data and llms.txt can be useful parts of a website's technical organisation, but Perplexity does not document either as a guaranteed citation-ranking factor. 6. Get your website discovered through traditional search Don't abandon traditional SEO in favour of AI search optimisation. I recommend: Set up Google Search Console. Set up Bing Webmaster Tools. Submit your XML sitemap. Ensure important pages are indexed. Research relevant search queries. Earn relevant, legitimate backlinks. Fix duplicate content and broken links. Search visibility and external references can create additional discovery opportunities. 7. Test whether Perplexity actually cites your website Don't rely on assumptions. Test actual queries. For example, if your website sells accounting software, search Perplexity for: Best accounting software for small businesses. How does accounting automation work? What features should accounting software include? Alternatives to traditional accounting software. Record: Whether your website appears. Which page gets cited. Which competitors get cited. Which questions generate citations. Whether citations remain consistent over time. Repeat the tests periodically and record the results. A practical 30-day implementation plan Week 1 – Technical foundations Configure robots.txt. Check PerplexityBot access. Check firewall restrictions. Set up Google Search Console and Bing Webmaster Tools. Submit your sitemap. Week 2 – Content development Identify 20 relevant customer questions. Create 5–8 comprehensive articles. Add supporting evidence and references. Improve internal linking. Week 3 – Authority building Publish original research or useful resources. Seek relevant industry coverage. Build relationships with reputable publishers. Improve your author and company information. Week 4 – Testing and improvement Test 20–30 relevant Perplexity questions. Record citations and competing sources. Identify content gaps. Improve pages that aren't being discovered. Common misconceptions Claim Reality Installing an llms.txt guarantees citations No Adding FAQ schema guarantees citations No Publishing hundreds of articles guarantees visibility No Allowing PerplexityBot guarantees citations No Paying for backlinks guarantees citations No Quality, accessible content improves your opportunity Sensible strategy, but no guarantee Perplexity's own documentation is the best starting point: Perplexity crawler documentation. Perplexity My recommendation If you're launching a completely new website, I'd prioritise these five things: Technical accessibility: Make sure Perplexity can crawl your website. Question-focused content: Publish genuinely useful answers to specific questions. Topical authority: Become a credible source within one particular subject. External recognition: Build a genuine reputation beyond your own website. Continuous testing: Measure actual Perplexity citations rather than vanity metrics. One final question: Are you trying to get a new website cited by Perplexity for a business, an e-commerce store, a SaaS product, or a content/blog website? The strategy differs considerably depending on the website type. I can put together a more targeted Perplexity SEO and AI visibility strategy, including a content plan and technical checklist.
1 recorded sources
What this observation cannot show
Authenticated free-account ChatGPT browser run in a fresh conversation. The product did not expose its complete retrieval configuration. The source panel reported 1 sources and 1 unique source records were captured. Results can vary by session, locale, account state and date.