What is Perplexity?
Perplexity AI
Perplexity is an AI answer engine that searches the web for almost every question and shows numbered citations next to its answer. That makes its sources unusually easy to analyse, so you can see exactly which pages shape what it says about your brand.
Perplexity is an AI answer engine that searches the web and cites its sources.
Its numbered citations show which pages shape each answer.
Being retrievable and quotable matters more in Perplexity than training data alone.
How Perplexity works
Perplexity works by running a web search for your question, reading the most relevant pages and writing an answer with numbered citations that link to those pages. The company, based in San Francisco, launched its answer engine in December 2022.
A typical Perplexity answer includes:
The answer: a written response, with numbers in the text that point to sources.
Sources: the pages it used, listed with their titles and domains.
Follow-up questions: suggested next questions to continue the conversation.
Paid Perplexity Pro users can choose which underlying model writes the answer, including models from OpenAI, Anthropic and Google, as well as Perplexity's own Sonar models. Perplexity has also launched Comet, an AI browser.
Because Perplexity retrieves on almost every query, what it says depends heavily on which pages it finds, a pattern known as retrieval-augmented generation.
Why Perplexity matters for AI visibility
Perplexity matters for AI visibility because it shows its work. Many assistants answer from training data or show only a few links. Perplexity cites sources for nearly every answer, so you can trace why a brand was or was not mentioned.
That makes it useful as a diagnostic engine. If Perplexity recommends a competitor, you can open the cited pages and see what they say. Often the answer comes from a comparison article, a review site or a forum thread, not from the competitor's own website.
It also means change can show up faster. Because Perplexity searches the web at answer time, a new page or a new mention on a cited site can appear in answers once Perplexity finds it. You do not have to wait for a model to be retrained, as you do with parametric knowledge.
Perplexity vs ChatGPT
Perplexity and ChatGPT both answer questions with AI, but they started from different places. Perplexity was built as a search product. ChatGPT was built as a chat assistant and added ChatGPT search later.
Retrieval: Perplexity searches the web for almost every question, while ChatGPT often answers from its training and searches when needed.
Citations: Perplexity shows numbered citations by default, while ChatGPT shows sources mainly when it searches.
Models: Perplexity lets Pro users choose among several model providers, while ChatGPT runs on OpenAI's models.
Use: Perplexity is used heavily for research, while ChatGPT covers a wider mix of tasks.
The same prompt can produce different brands in each. Track both rather than assuming one represents the other.
How to improve your visibility in Perplexity
Improving your visibility in Perplexity means being easy to find, easy to quote and present on the sites it already cites.
Allow its crawlers: Perplexity publishes the user agents and IP ranges of its crawlers, so check them against your robots.txt and firewall rules.
Answer specific questions: give each question a clear heading and a direct answer in the first sentence.
Study the cited sources: see which domains appear for your prompts and earn coverage there.
Keep pages current: update facts, prices and dates that engines may quote.
Track citations, not only mentions: see which of your pages are used and for which prompts.
Some publishers and Cloudflare have accused Perplexity of crawling in ways that bypass blocks, which Perplexity has disputed. If crawler access matters to you, check your server logs rather than relying on robots.txt alone.
How KIME tracks your brand in Perplexity
Perplexity is one of the ten AI engines KIME tracks. In Track & Analyze, you run your prompts in Perplexity alongside ChatGPT, Gemini, Copilot and the other engines, by country and language. You can see whether you are mentioned, where you appear and which domains and URLs Perplexity cites, then compare that with competitors to find the sources worth targeting.
Frequently asked questions
Is Perplexity a search engine?
Perplexity calls itself an answer engine. Instead of returning a list of links, it searches the web, reads the most relevant pages and writes an answer with numbered citations. You can still click through to every source. That makes it closer to a search engine than most chat assistants, while giving you a written answer rather than ranked results.
How does Perplexity choose its sources?
Perplexity runs web searches for your question and selects pages that look relevant and useful for the answer. It does not publish a full ranking formula. In practice, cited pages tend to answer the question directly and clearly. Tracking which domains it cites for your prompts is the most reliable way to learn what works in your category.
How do I get my website cited by Perplexity?
Make sure Perplexity's crawlers can reach your pages, then publish pages that answer specific questions in clear, quotable sentences. Earn mentions on the review sites, publications and forums Perplexity already cites for your topic. Keep facts current. Track your prompts over time to see which of your pages start appearing as citations.