GEO Brand Monitor

Monitor your brand's visibility in ChatGPT Search results. This Actor checks two layers of brand presence:

1. Direct AI citation - Does ChatGPT Search mention your brand in its response?
2. Source-level presence - Do the third-party sources that ChatGPT cites also mention your brand?

Sources that don't mention your brand are outreach opportunities. For those, the Actor scrapes contact details you can use for content placement outreach.

How it works

For each search query you provide, the Actor:

1. Queries ChatGPT Search via Google Search Scraper to get the AI-generated response and cited sources.
2. Checks brand mention in the AI response using case-insensitive text matching.
3. Filters sources by removing your own domains and competitor domains, leaving only third-party sources.
4. Crawls each source page using Website Content Crawler.
5. Checks brand mention in each source using case-insensitive text matching.
6. Scrapes contacts from sources that don't mention your brand (outreach opportunities) using Contact Details Scraper.

Input

Field	Type	Description
queries	string (textarea)	One search query per line
brand	string	Brand name to search for (case-insensitive)
ownDomains	string[]	Your domains to exclude from analysis
competitorDomains	string[]	Competitor domains to exclude
maxContactsPerDomain	integer	Max contact leads per source domain (default: 3)
maxSourcesPerQuery	integer	Max sources to process per query (default: 10)

Example input

{
    "queries": "best web scraping tools\nweb scraping platforms\nhow to scrape websites",
    "brand": "Apify",
    "ownDomains": ["apify.com", "use-apify.com"],
    "competitorDomains": [
        "brightdata.com", "oxylabs.io", "scrapingbee.com", "scraperapi.com",
        "zyte.com", "octoparse.com", "parsehub.com", "import.io",
        "webscraper.io", "diffbot.com", "phantombuster.com", "crawlbase.com",
        "thunderbit.com", "browserbase.com"
    ],
    "maxContactsPerDomain": 3,
    "maxSourcesPerQuery": 10
}

Output

Each row in the output dataset represents one source per query:

Field	Description
query	The search query
aiResponse	Full ChatGPT Search response text
brandMentionedInAI	Whether the brand appears in the AI response: Yes, No, or Error
sourceUrl	URL of the cited source
sourceDomain	Domain of the source
brandMentionedInSource	Whether the brand appears in the source content: Yes, No, or Error
contacts	Array of contact details (only for outreach opportunities)

Example output

{
    "query": "best web scraping tools",
    "aiResponse": "The full ChatGPT Search response text...",
    "brandMentionedInAI": "Yes",
    "sourceUrl": "https://example.com/article",
    "sourceDomain": "example.com",
    "brandMentionedInSource": "No",
    "contacts": [
        {
            "name": "John Doe",
            "email": "john@example.com",
            "position": "Editor"
        }
    ]
}

Use cases

• Generative Engine Optimization (GEO) - Track how your brand appears in AI-generated search results.
• Brand monitoring - Monitor which queries surface your brand in ChatGPT Search.
• Competitive analysis - Understand where competitors are mentioned and you're not.
• Outreach opportunities - Find third-party sites that cover your topic but don't mention your brand, along with contact details for outreach.

Actors used

This Actor orchestrates three Actors:

• Google Search Scraper - Fetches ChatGPT Search results
• Website Content Crawler - Crawls cited source pages
• Contact Details Scraper - Scrapes contact information from outreach opportunity domains

Cost considerations

• Google Search Scraper with ChatGPT Search is the base cost per query.
• Website Content Crawler runs once per source (controlled by maxSourcesPerQuery).
• Contact Details Scraper runs only for sources that don't mention your brand. Use maxContactsPerDomain to limit costs.

To minimize costs during testing, use 1-2 queries with maxSourcesPerQuery set to 3.