Perplexity Brand Monitor — Track AI Search Visibility

Know exactly where your brand stands in the AI search era. This actor queries the Perplexity AI API across any set of search prompts, detects brand mentions, measures share of voice across source citations, and outputs structured data ready for dashboards, spreadsheets, or CRM workflows.

Built for SEO agencies, brand managers, and CMOs who need to measure and improve their AI search visibility — also called GEO (Generative Engine Optimization) or Answer Engine Optimization (AEO).

---

Why AI Brand Monitoring Matters

Perplexity AI now handles hundreds of millions of queries per month. When a potential customer asks "what's the best CRM for a startup?" or "top project management software for remote teams?", the answer they receive — and which brands get cited — directly impacts purchase intent. Traditional rank tracking tools don't capture this. This actor does.

---

What It Does

• Queries Perplexity AI with your chosen prompts and checks each answer for brand mentions across your full competitive set
• Measures share of voice as a percentage of source citations — the core metric for AI search optimization
• Records citation position, mention count, and answer snippets so you know exactly how Perplexity presents your brand
• Supports sonar and sonar-pro models for different depth/cost tradeoffs
• Exports clean, structured data to Apify's dataset storage for use in Google Sheets, Airtable, Zapier, Power BI, or Looker Studio

---

Why Use This Instead of Dedicated Tools

Dedicated AI brand visibility platforms charge significant monthly fees before you've proven ROI:

Tool	Starting Price
Otterly.ai	$29–$489/mo
Profound	$499/mo
This actor (Apify PPE)	Pay only per result produced

This actor talks directly to the Perplexity API — no browser automation, no scraping, no anti-bot risk. Results are reliable and reproducible. You pay only for the Perplexity API calls you make (priced per token by Perplexity) plus a small Apify compute fee per actor run.

---

How to Get a Perplexity API Key

1. Go to https://www.perplexity.ai/settings/api and sign in or create an account
2. Click Generate to create a new API key
3. Copy the key and paste it into the Perplexity API Key field in this actor's input

Perplexity charges per token for API usage. Running 10 prompts uses roughly the same token count as 10 short chat conversations.

---

How It Works

1. Validate input — checks that an API key, at least one brand, and at least one prompt are provided
2. Iterate prompts — loops through each prompt (up to maxPrompts), sending each to the chosen Perplexity model via the chat completions API with a 1-second delay between calls to respect rate limits
3. Parse the response — extracts the answer text and the list of source URLs Perplexity cited to support its answer
4. Analyze each brand — for every brand in your list, counts how many times the brand name appears in the text, checks whether the brand's domain appears in any citation URL, records citation position, and calculates share of voice as (brand citations / total citations) × 100
5. Push results — writes one dataset record per brand-per-prompt combination, so 3 brands × 10 prompts = 30 rows

---

Input Parameters

Parameter	Type	Required	Description
perplexityApiKey	string	Yes	Your Perplexity API key (stored securely, never logged)
brands	array	Yes	Brand names to monitor — e.g. ["HubSpot (hubspot.com)", "Salesforce (salesforce.com)"]. Append domain in parentheses to also track domain citations.
prompts	array	Yes	Search prompts to run — e.g. ["best CRM software 2026", "top project management tools for remote teams"]
maxPrompts	integer	No	Maximum prompts to process in this run (default: 10, max: 100). Use 3 for testing.
model	string	No	Perplexity model: sonar (fast, cost-effective) or sonar-pro (deeper research, more citations). Default: sonar.

Choosing good prompts: Use the queries your target customers actually type into Perplexity. Think bottom-of-funnel comparison queries ("best [category] software", "top [category] tools for [use case]", "[your brand] vs [competitor]") rather than broad informational queries.

Tracking competitors: Add competitor brand names alongside your own. Each prompt is checked against every brand independently, giving you a full competitive picture per query.

Domain citation tracking: Format brands as "Brand Name (domain.com)" to track both text mentions and domain citations. Without a domain, only text mentions are tracked.

---

Output Fields

Each row represents one brand checked against one prompt.

Field	Type	Example	Description
prompt	string	"best CRM software 2026"	The search prompt sent to Perplexity
brand	string	"HubSpot"	Brand name tracked
mentioned	boolean	true	Whether the brand appeared in the answer text
mentionCount	integer	3	Number of times the brand name appeared in the answer
citedInSources	boolean	true	Whether the brand's domain appeared in Perplexity's source list
sourcePosition	integer	2	Position of the first brand citation (0 if not cited)
totalSources	integer	8	Total source citations returned in the answer
shareOfVoice	number	12.5	Percentage of citations that belong to this brand (0–100)
answerSnippet	string	"HubSpot is widely..."	Excerpt from the answer text around the first brand mention
allSources	array	["https://hubspot.com/crm", ...]	All source URLs Perplexity cited
checkedAt	string	"2026-03-19T14:22:00.000Z"	ISO 8601 timestamp of the check

Sample Output Record

{
  "prompt": "best CRM tools for small businesses",
  "brand": "HubSpot",
  "mentioned": true,
  "mentionCount": 3,
  "citedInSources": true,
  "sourcePosition": 2,
  "totalSources": 8,
  "shareOfVoice": 37.5,
  "answerSnippet": "...HubSpot is one of the most popular CRM tools for small businesses, offering a generous free tier...",
  "allSources": [
    "https://hubspot.com/crm",
    "https://g2.com/categories/crm",
    "https://forbes.com/advisor/business/best-crm-software"
  ],
  "checkedAt": "2026-03-20T09:15:00.000Z"
}

Share of voice is the most actionable metric: if your brand appears in 2 of 8 sources, your share of voice for that prompt is 25. Track this over time to measure the impact of your GEO and content efforts.

---

Use Cases

Weekly brand visibility report Schedule this actor to run every Monday with 20–30 high-intent prompts relevant to your category. Export the dataset to Google Sheets via Apify's Google Sheets integration. Track shareOfVoice and mentioned week over week to spot trends and measure content ROI.

Competitive AI search audit Before a product launch or repositioning, run a one-time audit with 50+ prompts covering your full competitive landscape. Use citedInSources and sourcePosition to identify which prompts competitors dominate and where you have gaps.

GEO content gap analysis For each prompt where mentioned is false, you have a content gap. The allSources array tells you exactly which pages Perplexity currently favors — audit those pages to understand what content signals drive AI citations, then create competing content.

Agency client reporting Run one actor instance per client with their brand plus top 3–5 competitors. Each run produces a clean dataset you can pull into a Looker Studio or Power BI report showing exactly how each client stacks up in AI search.

PR campaign measurement Run a baseline before a major press push, then re-run two to four weeks later. The shareOfVoice delta across high-volume prompts shows whether your earned media translated into AI search visibility.

Brand reputation monitoring Set up prompts like "[your brand] reviews", "is [your brand] trustworthy", "[your brand] alternatives". Monitor the answerSnippet field to see how Perplexity describes your brand in response to reputation queries.

---

Scheduling and Automation

Apify's built-in scheduler lets you run this actor on any cron schedule without writing code. Recommended cadences:

• Daily — fast-moving competitive categories or active PR periods
• Weekly — standard ongoing brand monitoring for most use cases
• Monthly — baseline benchmarking alongside traditional SEO rank tracking

Combine with Apify's Webhooks to push results to Slack, Notion, or any HTTP endpoint the moment a run completes.

---

Pricing

This actor uses Apify's pay-per-event (PPE) pricing — a small fee per dataset row written. Running 10 prompts against 3 brands produces 30 rows.

You also pay Perplexity directly for API token usage. See Perplexity pricing for current rates. The sonar model is the most cost-effective choice for routine monitoring.

There are no monthly subscriptions or seat fees.

---

Use via Apify API

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });

const run = await client.actor('samstorm/perplexity-brand-monitor').call({
    perplexityApiKey: 'YOUR_PERPLEXITY_API_KEY',
    brands: ['HubSpot (hubspot.com)', 'Salesforce (salesforce.com)'],
    prompts: ['best CRM software 2026', 'top CRM for small businesses'],
    maxPrompts: 10,
    model: 'sonar',
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

---

FAQ

Do I need a Perplexity subscription? No — you need a Perplexity API key, which is separate from a Perplexity Pro subscription. Sign up at perplexity.ai/settings/api. API usage is billed per token.

How many prompts should I use? Start with 5–10 prompts targeting your highest-priority queries to validate the setup. Scale up to 30–50 for comprehensive monitoring. maxPrompts caps the run to prevent accidental overuse during testing.

Why is my brand not detected even though I see it in the answer? Brand detection uses word-boundary matching. If Perplexity mentions your brand as part of a URL or with punctuation attached, it may not match. Try adding variant spellings as separate brand entries (e.g., "HubSpot" and "Hubspot").

Can I track the same brand with multiple domains? Yes — add it as two separate entries: "HubSpot (hubspot.com)" and "HubSpot (hs-sites.com)". Both will be checked and produce separate rows.

Is this scraping Perplexity's website? No. This actor uses Perplexity's official API. There is no browser automation, no scraping, and no risk of IP bans or bot detection.

How is share of voice calculated? shareOfVoice = (number of citations containing your brand domain / total citations) × 100. If Perplexity returns 8 sources and 2 are your domain, your share of voice is 25.0.

Can I monitor multiple clients or brands in one run? Yes — add all brands to the brands array. Each brand is checked independently against every prompt. For clean per-client data isolation, run separate actor instances per client.

Does the answer change every time I run the same prompt? Perplexity's answers are non-deterministic — the same prompt may return slightly different answers and sources on different runs. This is expected behavior and is why trend analysis over multiple runs is more meaningful than a single snapshot.

---

Limitations

• Perplexity API dependency — requires a valid Perplexity API key; fails if the API is unavailable or rate-limits your key
• Non-deterministic results — the same prompt may return different sources across runs; track trends rather than individual data points
• String-based brand detection — matches brand names by exact string with word boundaries; abbreviations, misspellings, or emoji variants are not detected unless added as separate entries
• No historical backfill — the actor captures a snapshot at run time only; there is no way to retrieve past Perplexity answers
• Domain matching only — citation tracking requires a domain; if your brand is cited via a third-party review site (e.g., G2 or Forbes), it counts as a competitor citation, not yours

---

Related Keywords

Perplexity AI brand monitoring, AI search visibility tracker, brand mentions in AI search, Perplexity scraper, GEO generative engine optimization, answer engine optimization, AI citation tracking, share of voice AI, Perplexity API, track brand Perplexity, AI SEO tool, brand visibility AI search, competitive intelligence AI.