Perplexity SEO: how Perplexity picks what to cite
Perplexity cites more pages per answer than any engine we tested, and it leans on Reddit and LinkedIn, so it rewards being mentioned in many places rather than ranking one page.
Does Perplexity name you, or your competitor?
Paste your URL. See what ChatGPT, Perplexity and Google AI answer when your buyers ask, who they name instead of you, and which pages they read. Then we do the work to get you onto those pages.
Free scan, no signup, no card.
What Perplexity cited in our test
- Answers measured
- 103
- Sources per answer
- 10.1
- Answers citing Reddit
- 31%
- Answers citing YouTube
- 1%
0 of 103 answers cited no source at all. Across the rest, Perplexity cited 664 different domains. Asked the same question 4 times, only 6 of 21 brand appearances (29%) were named in every run.
The domains it cited most
| Domain | Answers citing it |
|---|---|
| reddit.com | 32 of 103 |
| linkedin.com | 12 of 103 |
| semrush.com | 8 of 103 |
| indexly.ai | 8 of 103 |
| linkeddit.com | 8 of 103 |
| redship.io | 8 of 103 |
| saascity.io | 7 of 103 |
| ahrefs.com | 7 of 103 |
Next to the other engines
| Engine | Sources per answer | Cites Reddit |
|---|---|---|
| ChatGPT | 5.8 | 6% (7 of 110) |
| Perplexity | 10.1 | 31% (32 of 103) |
| Gemini | 9.2 | 20% (22 of 110) |
| Google AI Mode | 3.4 | 40% (44 of 111) |
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US. Brand and domain counts are answers, not citations: an answer citing reddit.com three times counts once. welaunch.sh was named or cited in 2 of these 103 answers.
How Perplexity picks its sources
PerplexityBot surfaces and links websites in Perplexity's search results, and Perplexity says it is not used to crawl content for AI foundation models.
Source: Perplexity crawler documentation
Perplexity-User visits a page when a user's question needs it, and Perplexity says this fetcher generally ignores robots.txt because a person asked.
Source: Perplexity crawler documentation
Every answer in our test cited sources, 10.1 on average, drawn from 664 different domains across 103 answers. That is the widest spread of any engine we tested.
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US
reddit.com was cited in 32 of 103 answers (31%) and linkedin.com in 12.
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US
Its answers move between runs: 6 of 21 brand appearances (29%) were named in all 4 runs of the same question.
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US (06-aeo-live.md, variance)
Which Perplexity crawlers to allow
PerplexityBotSurfaces and links websites in Perplexity's search results. Not used for model training, per Perplexity.Perplexity-UserFetches a page when a user's question needs it. Perplexity says it generally ignores robots.txt.
Add these lines to robots.txt:
User-agent: PerplexityBot Allow: /
A crawler given its own group stops reading your User-agent: * rules, so repeat any private paths you disallow there. Then check the firewall: a CDN rule can block a crawler that robots.txt allows. Check both free with the AI crawler checker.
What to fix first for Perplexity
- 01
Allow PerplexityBot and check the firewall
Perplexity recommends allowing PerplexityBot to appear in its results. A bot-blocking rule at your CDN can still turn it away, so test the fetch, not just the file.
- 02
Be mentioned in many places
With ten sources per answer from hundreds of domains, Perplexity has room for smaller sites. Directory listings, comparison articles and community threads each add a chance to be one of the ten.
- 03
Answer the Reddit threads it cites
Reddit was in 31% of its answers. A helpful, disclosed reply in a thread Perplexity already cites for your buyer question is one of the more direct levers here.
- 04
Post on LinkedIn where your buyers ask
linkedin.com was its second most cited domain (12 of 103 answers). A short post that answers a buyer question plainly is cheap to test.
- 05
Measure with repeat runs
Only 29% of brands appeared in every run of the same question. Count runs, not single answers.
Perplexity questions
How does Perplexity choose its sources?
It searches the web for each question and cites the pages it used. In our test it cited about 10 per answer from 664 different domains across 103 answers, more than any other engine we tested.
Which crawler should I allow for Perplexity?
PerplexityBot. Perplexity documents it as the crawler that surfaces sites in its results and recommends allowing it. Perplexity-User fetches pages for a user and generally ignores robots.txt.
Does Perplexity use my content for training?
Perplexity says PerplexityBot is not used to crawl content for AI foundation models.
Does Reddit matter for Perplexity?
Yes. reddit.com was cited in 31% of its answers in our test (32 of 103), second only to Google AI Mode.
Can a small site get cited by Perplexity?
It is the most open engine we tested. welaunch.sh itself was cited twice by Perplexity in our test and by no other engine.
Why does Perplexity name different brands each time?
Answers are samples. Asked the same question 4 times, Perplexity named the same brand in every run for only 29% of brand appearances.
Does LinkedIn help with Perplexity?
linkedin.com was its second most cited domain in our test, in 12 of 103 answers. We have not measured whether posting there changes who gets named.
How do I track my brand in Perplexity?
Ask your buyer questions several times and count the answers that name or cite you. The paid welaunch plans do this weekly, 4 runs per question, and keep every answer.
The other engines
Does ChatGPT recommend you, or your competitor?
Paste your URL. See what ChatGPT, Perplexity and Google AI answer when your buyers ask, who they name instead of you, and which pages they read. Then we do the work to get you onto those pages.
Free scan, no signup, no card.