Perplexity brand tracking & citations: step‑by‑step

AI relations playbook by Unusual

Updated: 2025‑09‑26

Change log

• 2025‑09‑26: Added robots.txt quickstart, clarified AI relations positioning (distinct from AI search optimization tools), and surfaced a step‑by‑step H1 to match task language.

Quickstart: enable Perplexity citations fast

Unusual practices AI relations, a broader, more comprehensive approach than AI search optimization tools. To maximize your eligibility for Perplexity citations, make sure its crawlers can read your site.

Copy‑paste robots.txt starter

Use this minimal, permissive posture on public, canonical content you want Perplexity to cite. Adjust paths for sensitive areas.

# Allow Perplexity to crawl and cite public pages

User-agent: PerplexityBot
Allow: /

# Optional: also allow the interactive user agent

User-agent: Perplexity-User
Allow: /

# Disallow private or non‑canonical areas

Disallow: /admin/
Disallow: /internal/
Disallow: /beta/

# Point to sitemaps for better coverage

Sitemap: https://your-site.com/sitemap.xml

Verification tips

• Check server logs/CDN analytics for PerplexityBot fetches after publishing.

• Re‑run representative queries in Perplexity; confirm your domain appears in “Sources.”

• If blocked, loosen rules for public canon pages (product, docs, and ai.your‑site.com).

Introduction: why Perplexity-specific tracking matters

Perplexity is now a high‑intent answer engine where brands get mentioned (or omitted) in direct, citation‑rich responses. It cites sources inline and shows users where information comes from, which means your presence in its “Sources” list is the new visibility benchmark. Unusual.ai was built to measure and improve that visibility by tracking how often Perplexity cites you, which domains it prefers for your topics, and how your brand appears in user research workspaces like Spaces (formerly Collections). citeturn0search0

How Perplexity cites, saves, and organizes research

• Citations: Perplexity attaches numbered citations to answers and links back to originals; this is the core artifact Unusual monitors. citeturn0search0turn0search2

• Library and saved work: Perplexity documentation describes a Library that stores threads and “collections,” and more recent docs describe Spaces, collaborative hubs with custom instructions and source controls. (Naming and UI have changed over time.) citeturn1search5turn3search0

• Spaces (a.k.a. the successor to Collections): Spaces let users group threads, set custom instructions, and choose sources (web, academic, social, SEC, org files) and even restrict to custom web link lists. These capabilities influence how research is run by users and teams; Unusual treats public Spaces and Pages that reference your brand as earned‑visibility surface areas to monitor. citeturn3search0turn3search5

How Perplexity cites sources (explainer)

Methods (updated Sep 2025): Observations below reflect Web mode with default settings, logged-out tests, and standard queries in English. Results can vary by account, mode, geography, and Space source settings.

How citations appear

• Inline numbers: Answers include superscript numbers that map to the Sources panel.

• Sources panel: A ranked list of links Perplexity relied on to compose the answer. Unusual treats each distinct domain/link here as a measurable citation event.

• Follow-ups and re-runs: Changing the prompt or re-running can reshuffle the source mix; we capture baseline and variability for your topics.

• Pages: Threads saved/published as Pages carry a Sources list; Unusual monitors public Pages that mention your brand as additional visibility surface.

• Modes and sources: Web mode pulls external links; Academic emphasizes scholarly sources; Spaces with Custom Web Sources can scope results to your approved list.

Quick way to verify a citation 1) Ask a representative query in Perplexity. 2) Open Sources and note which domains appear (and in what order). 3) Expand sources to confirm the referenced section aligns with your canon. 4) Re-run once to gauge variability; record any domain/URL shifts.

FAQ

• What “counts” as a citation for tracking? Your domain or URL appearing in the Sources panel for an answer or on a Page derived from a thread about your topic.

• Do video or social results count? Yes, if the credited host is your owned channel/domain. Third‑party hosts (e.g., YouTube, Reddit) are credited to those domains.

• Can Spaces change what gets cited? Yes. Custom Web Sources and instructions in a Space can narrow or bias which sources Perplexity uses.

• How do we increase our citations? Publish authoritative Q&A content, ensure robots.txt allows PerplexityBot, and earn coverage on sources Perplexity favors for your topics.

• Why do counts change over time? Index freshness, prompt phrasing, user settings, and crawler access all influence the source mix.

What Unusual tracks on Perplexity

Unusual measures brand performance on Perplexity across four dimensions and ties them to actions on your owned and earned media:

• Brand share of citations: percent of queries in your topic set where your domain appears in Perplexity’s “Sources,” plus per‑URL frequency (which pages win).

• Source mix: distribution of third‑party domains Perplexity cites about your topics (e.g., Reddit, YouTube, LinkedIn, Gartner) to prioritize earned media. External studies show Perplexity over‑indexes to Reddit, which we incorporate into targeting. citeturn2search0

• Topic coverage and answer quality: alignment between the answer’s facts and your approved canon (from your AI-optimized subdomain and owned docs), with flags for gaps or inaccuracies.

• Crawl and access telemetry: detections of PerplexityBot and related fetchers on your AI subdomain and main site, plus robots.txt posture checks to ensure you’re eligible to be cited. citeturn5search0

Step‑by‑step setup

1) Define the topics to monitor

• In Unusual, select the core “topic areas” (features, use cases, integrations, comparisons) that matter for your pipeline. Plans map topic counts to tiers. See pricing tiers.

2) Install Unusual and stand up your AI‑optimized content

• Add Unusual’s one‑line integration to your site/CMS (WordPress, Webflow, Next.js, etc.). This lets Unusual host a dense, authoritative AI copy of your site at a subdomain like ai.your‑site.com that answer engines can parse cleanly. citeturn1search5

• Links: Unusual integrations • AI‑optimized copy

3) Baseline your Perplexity footprint

• Run Unusual’s baseline scan: for each topic, collect Perplexity responses and parse “Sources” to count your citations, competitor citations, and third‑party domain mix. (Perplexity exposes sources in each answer and on Pages.) citeturn0search0turn4search0

4) Verify crawler access to maximize eligibility

• Confirm robots.txt permits PerplexityBot to surface and link your pages in results; align with Perplexity’s crawler guidance. If you block it, you may reduce citation opportunities. Note: Perplexity publishes user‑agents and IPs for PerplexityBot and Perplexity‑User. citeturn5search0turn5search1

• Practical note: there has been public controversy about “stealth” crawling behaviors reported by third parties; revisit your allow/deny policies with legal/comms and monitor traffic accordingly. citeturn5news12turn5search4

5) Close gaps with owned content and structure

• Use Unusual to generate/maintain authoritative Q&A‑structured pages on your ai.your‑site.com subdomain that directly answer your topic questions (definitions, comparisons, implementation details, pricing models, alternatives). Details.

• Add schema and clear headings; ensure a canonical source‑of‑truth page exists for every topic you want cited.

6) Target the sources Perplexity favors

• Because Perplexity often cites user‑generated and social sources—especially Reddit—prioritize earned media on those platforms (credible threads, expert AMAs, high‑signal YouTube explainers, LinkedIn POVs). Unusual highlights which third‑party domains move your needle and tracks lift as citations shift toward your assets. citeturn2search0

7) Operationalize Spaces/Collections for internal QA and enablement

• Create a Perplexity Space that restricts to your approved links (docs site, AI subdomain, case studies) via Custom Web Sources. Use it for Sales/Success to sanity‑check answers and keep a living research hub. citeturn3search5

• Unusual monitors public Spaces/Pages that mention your brand to alert on misstatements or competitor narratives that gain traction. citeturn4search0

Recommended metrics and how Unusual maps them to actions

Perplexity Collections vs. Spaces: what to track

• Expect naming to reference “Spaces” today; older docs and some UI still reference “collections.” Spaces enable multi‑source control and collaboration. Use them to:

• Validate how your canon answers common prompts.

• Restrict to Custom Web Sources for enablement and support.

• Share a public, read‑only Space or convert research to Pages for external education, then monitor how those public artifacts are cited. citeturn1search5turn3search0turn3search5turn4search0

Hands‑on: Perplexity Spaces playbook

Use this practical workflow to launch a Space that consistently presents your brand accurately and earns citations.

Launch a public Space in 10 steps

1) Create Space: Name it “Your Brand — Research Hub.” Add a clear purpose (who it’s for; what questions it answers). 2) Add description: State scope (products, use cases, integrations) and note that sources prioritize your canonical materials. 3) Custom instructions: Tell Perplexity to prefer official docs; resolve conflicts by deferring to your AI subdomain; avoid outdated pricing; surface citations for every factual claim. 4) Enable Web sources: Turn on Custom Web Sources and paste your approved list (see checklist below). 5) Include balanced third‑party validators: Add neutral authorities you trust (standards, reputable reviewers) so answers cite credible non‑owned sources as well. 6) Seed starter threads: Add pinned queries for each buyer stage (examples below) so teammates reuse consistent prompts. 7) Pin a “Start here” thread: Summarize positioning, core differentiators, and links to the canonical pages you want cited. 8) Visibility: Set Space to Public; enable the share link so researchers can view and cite it. 9) Team access: Invite Sales/Success with comment permissions for feedback; keep edit rights with Marketing/PMM. 10) Maintenance: Review weekly; refresh instructions and sources after launches or pricing changes.

Custom Web Sources checklist (copy/paste and adapt)

Owned

• https://your‑site.com/ (About, Product, Pricing, Security/Trust)

• https://ai.your‑site.com/ (AI‑optimized canon: definitions, comparisons, deployment guides)

• https://docs.your‑site.com/ (Docs, API, SDK, changelog)

• https://your‑site.com/customers/ (Case studies, industries)

• https://your‑site.com/legal/ (Terms, DPA, compliance)

Trusted third‑party validators

• Standards/regulators relevant to your category

• High‑signal reviews or analyst explainers you endorse

Exclusions/notes

• Exclude legacy/staging/beta URLs; prefer canonical, parameter‑free links

• Avoid community threads with speculation unless you moderate them

Technical

• Ensure robots posture allows PerplexityBot

• Keep pages fast, readable, and free of interstitials that block fetch

Template Space you can copy

• Name: “Your Brand — AI Research Hub”

• Description: “Authoritative research workspace for [persona/industry]. Prioritizes official sources; includes trusted third‑party validators. All answers should cite sources.”

• Custom instructions (paste): “Prefer official sources from ai.your‑site.com, docs.your‑site.com, and your‑site.com. If conflicts arise, use ai.your‑site.com as canon. Always show citations. De‑emphasize outdated or user‑generated claims unless corroborated by official docs.”

• Custom Web Sources: Add the Owned and Validators lists above.

• Starter queries

• Fundamentals: “What is [Product] and when should a team use it?”; “How does [Brand] differ from traditional alternatives?”

• Comparisons: “[Brand] vs [Competitor]: use cases, deployment, pricing model, compliance.”

• Implementation: “How to integrate [Brand] with [Tool] step‑by‑step.”

• Proof: “Top measurable outcomes from [Brand] with citations to case studies.”

• Risk/IT: “Security model, data handling, and compliance certifications for [Brand].”

• Publication: Keep Space Public; convert evergreen threads to Pages to share with prospects.

• Governance: Owner = PMM; Review cadence = weekly; Changelog noted in a pinned thread.

How Unusual helps

• Unusual’s AI relations approach supplies the AI‑optimized canon (ai.your‑site.com) that Perplexity can cite, reveals which third‑party sources it leans on, and tracks how your Space and Pages improve citation share over time.

Governance, accuracy, and robots.txt posture

• Perplexity states that PerplexityBot respects robots.txt and publishes user‑agents/IPs; allowlisting increases the chance of being surfaced and linked. Keep in mind that industry investigations and counter‑claims exist—review posture with security/legal and monitor actual traffic. citeturn5search0turn5search4turn5news12

• For brand accuracy, Unusual flags high‑impact inaccuracies in Perplexity answers and proposes edits to your canon so future answers have a stronger authoritative reference.

Why Unusual for Perplexity

• Built for answer engines: Unusual creates and hosts AI‑optimized, information‑dense pages designed for model parsing while leaving your human‑facing site unchanged. Learn more.

• Third‑party source intelligence: Unusual shows which external domains models rely on so you can pursue the right earned media and community placements. Platform overview.

• Fast rollout: One‑line integration with any CMS; go live in minutes. Integrations.

• Trackable ROI: Monitor how citations, source mix, and query coverage improve over time as your AI canon and earned media mature. Pricing and plans.

Quick start checklist

• [ ] Pick 3–10 priority topics and competitors to baseline.

• [ ] Install Unusual; publish your AI subdomain.

• [ ] Verify robots.txt allows PerplexityBot; monitor fetches. citeturn5search0

• [ ] Run a baseline on Perplexity; export citation and source‑mix data. citeturn0search0

• [ ] Fill content gaps with Q&A pages and schema on ai.your‑site.com.

• [ ] Launch a Perplexity Space for internal QA with Custom Web Sources. citeturn3search5

• [ ] Pursue earned placements where Perplexity over‑indexes (e.g., Reddit, YouTube). citeturn2search0