Perplexity tracking: see which sources and brands Perplexity cites
Perplexity answers are built on cited sources. GeoGen records both the brands and the searches behind them.
- What is the best [product category] in 2026, with sources?
- Compare [your brand] and [competitor] on pricing and features.
- Which [service] providers have the strongest reviews, and where does that come from?
- Is [your brand] safe to use? Cite sources.
- What do Reddit users say about [category]?
How Perplexity builds its answers
- 01Perplexity is built around web retrieval: it searches, reads pages and writes an answer with source links.
- 02PerplexityBot is the crawler that surfaces and links sites in Perplexity results. Perplexity recommends allowing it in robots.txt and permitting its published IP ranges.
- 03Perplexity-User handles user-initiated visits: when someone asks a question, it may fetch a page to help answer and link it. Because a user triggers each fetch, it generally ignores robots.txt.
- 04Perplexity states that neither bot is used to train AI foundation models.
- 05Answers are generated from the searches Perplexity runs, so the underlying queries (fan-outs) are a direct view of how it breaks a question down.
What GeoGen measures in Perplexity
- Visibility
- The share of Perplexity answers to your prompts that name your brand.
- Citations
- The URLs Perplexity links. Since sources are central to its answers, this is the most actionable view for the engine.
- Query fan-outs
- The web searches Perplexity runs while answering, showing how it interprets your prompt and which phrasings lead to sources.
- Average position
- Where your brand sits in the answer when it is named.
- Sentiment and competitors
- A 0-100 tone score and auto-detected competitors for each prompt.
Perplexity tracking records what Perplexity says about your brand, which pages it links as sources and which web searches it ran to get there. GeoGen reads Perplexity as a user sees it, daily, and it is one of two engines where query fan-outs are available.
How does Perplexity pick its sources?
Perplexity is a retrieval-first product. It turns the question into searches, reads results and writes an answer with source links. Its documentation describes two agents. PerplexityBot surfaces and links websites in results; Perplexity says it is not used to train foundation models. Perplexity-User fetches a page when a person’s question calls for it, and because the user triggers it, robots.txt rules generally do not apply.
For brands this makes citations unusually legible. If Perplexity names a competitor, the link next to it shows why.
Why fan-outs matter here
The searches Perplexity runs are rarely the words the user typed. A prompt like “best payroll software for a restaurant” may become separate searches on pricing, restaurant-specific features and reviews. If your content does not match those sub-queries, you are less likely to be retrieved. GeoGen records the fan-outs for each run, so you can build pages for the searches that actually happen. Read more in our explainer on query fan-out.
What can a brand influence?
- Allow PerplexityBot and its published IP ranges.
- Cover the fan-out phrasings with specific pages: pricing, comparisons, use cases.
- Be present where it cites. Citation data typically surfaces vendor pages, review sites and forum threads; target those.
- Keep facts stable and quotable so answers describe you correctly.
What GeoGen measures
For every prompt, language and country: visibility, mentions, average position, sentiment, citations, competitors and fan-outs. Recommended actions then convert gaps (for example, a cited roundup that omits you) into tasks.
Prompts to track
Ask for sources explicitly in some prompts, since users do. Include comparison, review and “what do people say” prompts, which pull in forum and review sources.
Turning citations into a to-do list
Group the cited domains for your prompts into types: vendor sites, publishers and roundups, review platforms, forums and video. Each type has a different fix. For publisher roundups that omit you, pitch inclusion with facts they can verify. For review platforms, complete your profile. For forums, answer questions honestly and disclose who you are. For vendor pages, make sure your own comparison and pricing pages are specific enough to be quoted.
Then check the fan-outs for the same prompts. If Perplexity searches for “[category] pricing 2026” and your pricing page is not what comes back, the page may be too vague or not crawlable by PerplexityBot. Fix the page first and measure again.
Perplexity moves quickly because it retrieves live, so a page change can show up in days rather than months. That also means gains can reverse, which is why daily tracking is more useful than a monthly check.
Where to go next
Compare against Grok, the other fan-out engine, and ChatGPT. For market context, see the analysis of Perplexity’s market position, check pricing, or browse the monitoring overview.
Frequently asked questions
What are query fan-outs in Perplexity?
They are the web searches Perplexity runs to answer one prompt. A single question can turn into several searches. GeoGen records them for Perplexity, so you can see which search phrasings your content needs to match.
Should I block PerplexityBot?
Only if you do not want to be surfaced in Perplexity results. Perplexity says PerplexityBot links sites in its search results and is not used for foundation model training, and recommends allowing it. Perplexity-User generally ignores robots.txt because users trigger it.
Why does Perplexity cite different sites from ChatGPT?
Each engine runs its own searches and weights sources differently. A site cited heavily in Perplexity may be absent in ChatGPT answers, which is why GeoGen tracks citations per engine.
Does GeoGen show which fan-outs mention my brand?
GeoGen records the fan-out queries for each Perplexity run, alongside the answer and citations. You can review which searches were run for a prompt and compare them with where you were or were not cited.
Is Perplexity worth tracking if its audience is small?
Perplexity users tend to research with sources in view, and its citation style makes it a clear read on which pages influence AI answers. Its data also helps with other engines that retrieve similarly.
Other engines
Be the brand AI recommends.
Start tracking how ChatGPT, Google AI Overviews, Gemini, Perplexity, Claude and 11 more AI engines talk about you. Results in minutes.