Start with access. ChatGPT and Perplexity each use a named crawler for search, and if your robots.txt blocks it, your pages cannot show up as sources. OpenAI says it uses "OAI-SearchBot" to surface websites "in search results in ChatGPT's search features" (OpenAI). Perplexity says "PerplexityBot is designed to surface and link websites in search results on Perplexity" (Perplexity). Once those bots can read your site, the rest is writing pages that answer one question clearly, with the facts and sources easy to lift.
Step 1: let the search crawlers in
Open yoursite.com/robots.txt in a browser. If you see a Disallow rule under User-agent: OAI-SearchBot, User-agent: PerplexityBot, or under User-agent: * covering your blog, those tools cannot use your pages.
OpenAI separates search from training. Its documentation describes allowing OAI-SearchBot "in order to appear in search results while disallowing GPTBot to indicate that crawled content should not be used for training" (OpenAI). So you can appear in ChatGPT search without letting your content train models. Perplexity notes PerplexityBot "is not used to crawl content for AI foundation models" (Perplexity).
A robots.txt that allows both search crawlers looks like this:
User-agent: OAI-SearchBot
Allow: /
User-agent: PerplexityBot
Allow: /
Check with your developer or site builder before changing it, especially if a security plugin manages the file for you.
Step 2: write pages an answer engine can quote
Nobody outside these companies knows exactly how they pick sources, so treat what follows as our working approach, not a guarantee.
- Put the answer in the first two sentences. If someone asks "how much does X cost," the opening should say the price range, not the history of X.
- One page, one question. A page titled "How long does a kitchen remodel take" is easier to cite for that question than a page covering everything about remodels.
- State facts in plain sentences. "The permit fee is $85 a year" can be quoted. A paragraph that circles the number cannot.
- Link your sources. A claim with a link to its origin is easier to trust and to check.
- Use tables for comparisons. Side-by-side prices or features are easy for a reader, and an answer engine, to pull from.
- Add an FAQ. Short questions with short answers match the way people phrase questions to chat tools.
- Say who you are and what you sell. Clear authorship and honest disclosures make a source safer to cite.
Step 3: be findable in regular search too
ChatGPT and Perplexity search the web when they answer, so pages that do well in normal search have a head start. In our view, the same work helps both: answer real questions people search, and keep publishing. If you want to know which questions to start with, your Google Search Console data shows the searches where you already almost rank. We wrote a step-by-step guide to finding them.
How Paperboy writes for this
Full disclosure: we make Paperboy, so weigh this accordingly. Every Paperboy article is written in the format above: the answer first, one question per article, each fact linked to the page it came from, a comparison table or checklist where it helps, an FAQ with FAQ and Article structured data, and a clear disclosure when a cited source sells something. We are building tracking of when AI tools mention your site; it is marked Coming soon in the app.
Frequently asked questions
Does blocking GPTBot stop my site from appearing in ChatGPT?
Not by itself. OpenAI uses OAI-SearchBot for ChatGPT search and GPTBot for training, and says you can allow OAI-SearchBot while disallowing GPTBot (OpenAI).
Is PerplexityBot used for AI training?
Perplexity says PerplexityBot is not used to crawl content for AI foundation models (Perplexity).
Can I guarantee ChatGPT will cite my page?
No. Nobody outside these companies controls which sources they pick. You can make your pages readable by their crawlers and easy to quote, which is what this guide covers.