Google Maps Scraper & Local Business Intelligence Agent

Turn Google Maps searches into decision-ready local business intelligence. This Actor extracts businesses from Google Maps, optionally enriches public website contacts, ranks commercial opportunities with a transparent deterministic score, compares current results with a prior snapshot, and emits compact fields that AI agents can act on directly.

Why use this Actor?

A conventional Google Maps scraper gives you rows. This Actor gives you rows plus a commercial decision layer: contactability, digital-opportunity score, reputation score, lead score, explainable evidence, recommended agent action, and exact field-level changes from a previous snapshot.

Key features

• Up to 100 Google Maps results per search query using a lightweight HTTP-first path.
• Multiple independent search queries in one run.
• Business name, stable Place ID, Maps URL, address, coordinates, categories, rating, review count, phone, and website.
• Optional website enrichment for public emails and social profiles.
• Optional DNS MX verification for discovered email domains.
• Deterministic lead scoring. No paid LLM is required to rank results.
• agentAction and agentReason fields designed for MCP/API/agent workflows.
• previousSnapshot comparison for changed phone, site, address, rating, reviews, categories, and name.
• Aggregate market summary and top leads stored in MARKET_SUMMARY.
• Current normalized rows stored in SNAPSHOT for the next monitoring run.

Example input

{
  "queries": ["dentists in Austin TX", "orthodontists in Austin TX"],
  "maxResultsPerQuery": 50,
  "enrichContacts": true,
  "verifyEmailDomains": true,
  "minRating": 3.5,
  "minReviews": 10,
  "leadObjective": "website redesign and local SEO outreach"
}

Output

Each dataset row contains raw business evidence and decision-ready fields such as leadScore, leadScoreExplanation, agentAction, agentReason, changes, and sourceConfidence. This keeps agent consumption compact while preserving the underlying evidence.

Change monitoring

Pass a previous SNAPSHOT array back as previousSnapshot. The Actor produces exact before/after changes and a bounded changeScore. This makes scheduled competitive monitoring possible without pretending that a normal one-shot scrape is historical intelligence.

Pricing

Pay per delivered business result. Platform compute is included in the Actor price. The initial price is intentionally positioned below the dominant full-featured Google Maps scraper while preserving room for contact enrichment and monitoring logic.

AI and automation readiness

The Actor is built for Limited Permissions, Pay-per-event pricing, API/OpenAPI access, scheduled runs, Tasks, datasets, and Apify's agentic discovery path. Compact decision fields reduce token waste for downstream agents.

Responsible use

Only public business information is collected. Website enrichment is limited to public pages. Users are responsible for lawful processing, outreach rules, privacy obligations, and applicable website terms.

Limitations

• Google Maps is an external surface and may change without notice.
• One query is currently capped at 100 returned businesses. Use multiple geographically specific queries for broader coverage.
• MX verification confirms that a domain can receive mail; it does not prove that a specific mailbox accepts mail.
• Lead scoring is an explainable prioritization heuristic, not a guarantee of conversion.
• Historical change intelligence requires a prior snapshot supplied by the user or a scheduled workflow that stores and reuses it.

Typical use cases

Local SEO prospecting, agency lead generation, competitive location monitoring, franchise research, reputation opportunity discovery, sales territory research, market mapping, and AI-agent business discovery.

Gen2 decision intelligence

This Actor preserves its original analysis and adds a decision layer with baseline awareness, regression detection, confidence, GO/WARN/BLOCK executive output, and an optional economic-impact estimate. Economic estimates are produced only when the user supplies valuePerImpactUnitUsd; the result states the calculation basis instead of inventing monetary value.