Google Maps B2B Lead Finder + Email Verifier + AI Enricher

Turn Google Maps search results into outreach-ready business leads. This Actor searches Google Maps for local businesses, crawls each business website for contact details, discovers and verifies email addresses, runs AI-based qualification, and outputs clean, scored lead objects ready for your CRM, outreach sequences, or downstream automation.

Extract local businesses from Google Maps, discover and verify contact channels, and let AI score which leads are worth contacting first.

What does this Actor do?

1. Searches Google Maps by keyword + location (or direct Maps URLs) and extracts place records: name, category, address, phone, website, rating, review count, coordinates, and open/closed status.
2. Deduplicates places found across multiple search terms (place ID, domain, phone, name+address keys).
3. Crawls each first-party business website (contact, about, and team pages first) and extracts emails (including mailto:, obfuscated, and JSON-LD sources), phone numbers, contact forms, booking links, and social profiles.
4. Verifies discovered emails - syntax, disposable-domain, and MX checks - using the standard verification vocabulary: ok, invalid, disposable, catch_all, unknown, error.
5. Runs AI enrichment (Anthropic, OpenAI, OpenRouter, or free built-in heuristics): niche classification, website quality score, digital maturity, opportunity tags, pain-point signals, and a one-line personalization hook.
6. Scores every lead 0-100 with an A/B/C/D quality band across contactability, commercial fit, digital-presence gap, reputation, and legitimacy.
7. Delivers results to the dataset (JSON/CSV/Excel export), a run-summary record, and an optional webhook.

Who is it for?

Lead generation agencies, local SEO and web design agencies, B2B sales teams, franchise expansion and market research teams, and AI workflow builders using n8n, Make, LangChain, or custom APIs.

Quick start

{
    "searchTerms": ["dentist", "dental clinic"],
    "locationQuery": "Malmö, Sweden",
    "maxPlaces": 200,
    "crawlWebsite": true,
    "verifyEmails": true,
    "enableAiEnrichment": true,
    "aiProvider": "none"
}

Set aiProvider to anthropic, openai, or openrouter with your own aiApiKey for LLM-grade enrichment, or keep none for free rule-based enrichment.

Output example

{
    "business": {
        "name": "Bright Smile Dental",
        "categoryName": "Dentist",
        "address": "Storgatan 1, 211 34 Malmö",
        "phone": "+46 40 123 456",
        "website": "https://brightsmile.se",
        "rating": 4.7,
        "reviewsCount": 132
    },
    "contacts": {
        "emails": [
            {
                "email": "info@brightsmile.se",
                "type": "generic",
                "sameDomain": true,
                "verification": { "status": "ok", "subResult": "mx_accepts_mail" }
            }
        ],
        "contactForms": ["https://brightsmile.se/contact"],
        "socials": { "facebook": ["https://facebook.com/brightsmile"] }
    },
    "ai": {
        "normalizedNiche": "family dental clinic",
        "digitalMaturityLevel": "medium",
        "websiteQualityScore": 62,
        "serviceOpportunityTags": ["needs-seo", "no-booking-flow"],
        "personalizationHook": "Family clinic with strong ratings but no online booking flow.",
        "leadQualityScore": 81
    },
    "scoring": {
        "leadQualityScore": 78,
        "leadQualityBand": "A",
        "contactabilityScore": 84
    }
}

Three dataset views are provided in the Output tab: Business leads, Contacts, and AI summary. A RUN_SUMMARY record with aggregate metrics is saved to the key-value store.

Pricing (pay per event)

You pay only for what the Actor actually produces:

Event	When it's charged
Place scraped	A business record passed your filters
Website crawled	A first-party website was successfully crawled
Same-domain email found	A unique same-domain email was discovered
Email verified (decisive)	Verification returned ok, invalid, or disposable - non-decisive results are free
AI enrichment produced	A valid AI enrichment result was produced
Qualified lead	A lead reached the A/B quality band

Base extraction stays cheap; you opt into website crawling, verification, and AI scoring only when you need them. Cost controls are built in: maxPlaces, maxWebsitePages, maxAiEnrichments, and minLeadQualityScore.

Input reference

See the Input tab for all options. Highlights:

Field	Purpose
searchTerms + locationQuery	Broad geographic discovery (recommended - covers more than a raw Maps URL)
startUrls	Direct Maps search/place URLs (note: Google caps a single search URL at ~120 results)
minRating, minReviews, skipClosed, categoryFilters	Maps-level filters
crawlWebsite, maxWebsitePages, contactPagePriority	Website crawl policy
discoverEmails, verifyEmails	Email discovery + verification
enableAiEnrichment, aiProvider, aiApiKey, aiPromptProfile	AI qualification
requireWebsite, requireEmailOrForm, minLeadQualityScore	Commercial usability filters
webhookUrl, batchId	Downstream automation

Example automations

• Push A-band leads into HubSpot/Pipedrive via the webhook + Make/n8n.
• Schedule weekly runs per territory using batchId to track batches.
• Chain with an outreach tool: filter dataset to verification.status = "ok" and export CSV.

Limitations

• Google Maps markup changes over time; the Actor uses defensive selectors and retries, but extraction coverage can vary. Report issues and we ship fixes.
• A single direct Maps URL search is capped by Google at roughly 120 results; use searchTerms + locationQuery for full coverage.
• Built-in email verification uses DNS/MX checks (no SMTP mailbox probe), so ok means the domain accepts mail, not a guaranteed mailbox.
• Use of proxies (proxyConfiguration with residential proxies) is strongly recommended for runs above ~100 places.

Legal & compliance

This Actor extracts publicly available business information (not private personal data). Scraping public data is generally legal, but you are responsible for how you use the output: respect GDPR/CAN-SPAM and regional outreach rules, honor takedown requests, and review the target sites' terms of use. Person-level extraction is off by default; the Actor practices data minimization (no raw HTML stored unless you opt in).