Open Perplexity. Ask it who’s best in your space. I’ll wait.
It answered, named a few brands, and cited its sources right there on the screen. Was one of them you?
If not, that’s the whole problem. And the good news: Perplexity is the easiest place to fix it.
Most engines hand you an answer and hide where it came from. Perplexity shows its working. Numbered sources, right next to the answer. You can see who got picked, who got skipped, and who your real competition is. That transparency is a cheat code, and almost nobody uses it.
It also runs live search instead of leaning on old training data. So it rewards fresh, specific, well-structured pages over big stale ones. A newer brand that answers one narrow question properly can beat a site that answered it vaguely two years ago. That is not true on most engines. It’s true here.
If you want the others, here’s how ChatGPT picks brands and why Claude plays by its own rules. Perplexity is different enough to deserve its own playbook.
First, can Perplexity even read you?
None of this matters if the crawler can’t see your page.
Perplexity crawls with PerplexityBot, and like every AI bot it reads the raw HTML your server sends back. It doesn’t click, scroll, or wait for JavaScript. If your answer only shows up after a tab or an accordion, it’s invisible.
Allow it:
User-agent: PerplexityBot
Allow: /
User-agent: GPTBot
Allow: /
User-agent: ClaudeBot
Allow: /Loads of sites quietly block AI bots through a security plugin or a Cloudflare rule and never notice. Run your domain through the robots.txt checker before anything else. Blocked means invisible. No exceptions.
Answer the question, not the keyword
Nobody types “best CRM” into Perplexity. They type “best CRM for a 15-person agency already using Slack.”
Long. Specific. Loaded with context. And wide open, because the big sites are all fighting over the short keyword.
So write the page that answers the whole question. Make your H2s read like the questions people actually ask, because Perplexity matches intent, not your meta tags. Want the fastest research shortcut? Ask Perplexity your ten real buyer questions and read the answers. It hands you the exact phrasing it rewards and the exact competitors you’re up against. Free intel, sitting in plain sight.
Say it first. Say it specific.
Perplexity lifts the cited line from the top of a section. So put the answer there. First sentence. No warm-up.
Then back it with something real. “Most teams see results in 60 to 90 days” gets cited. “Results vary” gets ignored. Give it a number, a range, a date, a named source, and it quotes you by name instead of paraphrasing you into a footnote. This answer-first habit is the whole basis of answer engine optimization, and it pays on every engine, not just this one.
Fresh beats big here
Perplexity favors recent content harder than any other major engine. It’s reading the live web, not a snapshot from last year.
So a page you updated last week can outrank a stronger page nobody has touched in months. Build the habit: revisit your important pages every one to three months and actually change something real. New data, a fresh example, a tighter answer. Not just the date. Perplexity can tell, and so can your readers.
Get talked about where Perplexity looks
This is the big one, and it’s the part you don’t control from your own site.
Perplexity leans hard on third-party mentions. A brand that keeps showing up on Reddit, review sites, and industry pages gets cited way more than a brand that only talks about itself. So the work lives off your site as much as on it: answer real questions in the communities it keeps citing, get listed where your category gets reviewed, and earn a mention in the pages that already rank as sources.
This is the same off-site engine behind generative engine optimization, and it’s where most of the citation weight actually sits. If you’d rather not run all this yourself, that’s literally the job when you hire an AEO consultant.
Small brand? Good. Win the edges.
If you’re new or small, don’t fight for the broad query. You’ll lose, for now.
Take the edges instead. Find the specific, current questions the big sites answered lazily or skipped. Be the sharpest, freshest, most concretely-sourced answer to that exact question. Stack a couple of targeted mentions behind it. Perplexity, reading live and rewarding relevance over size, will often pick the specific new page over the vague old giant. Do that across a cluster of related questions and you build a real footprint before your domain authority ever catches up.
That’s the whole reason to start here. Perplexity is the one engine where sharp and small beats big and lazy.
How to know it’s working
You don’t need a tool. You need a habit.
Once a month, ask Perplexity your fifteen most important buyer questions. Note which answers cite you, what text they pulled, and which competitors keep showing up. Date it. Repeat. The trend tells you plainly whether the crawl access, freshness, and off-site work are landing. Want a faster read across every engine? Run the AI Visibility Score.
Perplexity shows its work. Least you can do is show up in it.
Related reading
- How to get cited by ChatGPT: the Bing-based engine and its off-site rules
- How to get cited by Claude: why tone matters more on Anthropic’s engine
- Schema markup for AI search: the structured data that helps machines read you
- Generative engine optimization: the full multi-engine system
Want to know why Perplexity keeps naming your competitor and not you? Start with an audit and get a query-by-query breakdown of where you show up, where you don’t, and what’s blocking it.
Common questions,
direct answers.
Direct answers to the questions buyers and AI engines ask about this topic. Each answer is structured for citation in ChatGPT, Perplexity, and Google AI Overviews.
01 How does Perplexity choose which sources to cite?
Perplexity runs a live web search, then cites the pages that answer the question most directly, most recently, and most credibly, showing them on screen as numbered sources. It rewards a clean answer it can lift, specific data, fresh updates, and the same claim showing up on other trusted sites. Because the sources are visible, you can see exactly who it picks for any query.
02 How do I get my brand cited by Perplexity?
Let PerplexityBot crawl your raw HTML, answer the full question a buyer actually types, and lead each section with the answer plus a specific number. Keep the page fresh, and get mentioned on the third-party sites Perplexity already pulls from, like Reddit, review sites, and industry pages. Same answer-first foundation as AEO, tuned for live search.
03 Does Perplexity favor fresh content?
Yes, more than most engines, because it reads the live web instead of a frozen training set. A page updated last week can beat a stronger page that has sat untouched for a year. Revisit important pages every one to three months and actually update them, not just the date.
04 Is getting cited by Perplexity different from ChatGPT or Claude?
Same fundamentals, different emphasis. Perplexity leans harder on freshness and exact question matching, and it shows its sources openly, so you can see and reverse-engineer your competitors. ChatGPT and Claude hide more of their working, and Claude in particular cares more about tone.