Primelabs Google Maps Lead Extractor

A high-performance geocoding and spatial automation scraper built to harvest enriched commercial leads from Google Maps. By slicing target regions into an automated coordinate matrix and utilizing strict server-side tracking token extraction, this tool bypasses standard platform viewport limits and delivers full geographic market coverage while keeping your compute costs under control.

---

Why Use This Scraper

Full Market Coverage Standard scrapers search a city name and scroll. Google Maps caps that view at around 120 listings, which means most of the real market data never shows up. This actor splits the target area into a coordinate grid and runs independent searches across each cell, surfacing businesses in outer neighborhoods and suburbs that get missed entirely.

Strict Production Coordinate Accuracy Unlike public scrapers that capture drifted camera viewports or broad city centroids, this engine enforces a strict structural parsing pipeline. It extracts exact physical business coordinates directly from Google's canonical server-side !3d and !4d tracking tokens, preserving 100% data integrity.

Clean, Parameter-Isolated Deduplication Every profile undergoes structural isolation. The engine strips tracking arguments and query strings down to the unique !16s alphanumeric place_key. Duplicate listings across overlapping grid cells are intercepted and dropped before they can contaminate your CRM database.

---

Use Cases

Lead Generation Build prospecting lists for sales teams with verified phone numbers, addresses, and websites for any niche and location.

Local SEO and Web Design Outreach Use the built-in lead scoring system to find businesses with no website or a weak review profile, ideal targets for agency outreach.

Market and Competitor Mapping See how saturated a market is across an entire metro area, not just the downtown core, and benchmark businesses by rating, review volume, and online presence.

---

How the Grid Scan Works

The actor runs through three phases for every search.

Phase 1: Geocoding
Look up the target city and get its real boundary box

Phase 2: Grid Generation
Build an evenly spaced grid of coordinates across that boundary
[ Cell 1 ]   [ Cell 2 ]   [ Cell 3 ]
[ Cell 4 ]   [ Cell 5 ]   [ Cell 6 ]
[ Cell 7 ]   [ Cell 8 ]   [ Cell 9 ]

Phase 3: Extraction
Visit each cell at zoom level 16 and collect every visible listing

Searching each cell independently instead of running one broad city search is what allows the actor to reach suburbs and outer neighborhoods that normally never appear in results.

---

Production Lead Scoring

Every business that comes through gets a quality score from 0 to 100, calculated via a balanced multi-variable algorithmic formula:

Signal Matrix	Maximum Point Weight	Calculation Engine
Phone number present	25 Points	Binary confirmation check
Website present	25 Points	Link verification lookahead
Review Volume	25 Points	Volume calculation (reviews / 10.0, capped at 25)
Star Rating	25 Points	Direct scale multiplication (rating * 5.0, capped at 25)

90 to 100, Enterprise Tier Established business with verified contact details and strong review activity.

60 to 89, Growth Tier Active business with a clear gap, usually a missing website or low review count, good for outreach.

---

Output Schema

Each item in the dataset includes:

Field	Type	Description
record_id	String	Deterministic SHA-1 hash unique to the business identifier
place_key	String	Clean, parameter-stripped unique Google feature routing key
name	String	Business name as listed on Google Maps
business_status	String	Operating status, for example "operational" or "permanently_closed"
matched_keyword	String	The search term that returned this result
address	String	Full normalized street address with line-break stripping
phone	String	Cleaned, international format phone number
website	String	Cleaned destination URL listed on the business profile
total_rating	Float	Star rating from 0.0 to 5.0
review_count	Integer	Total number of public reviews
latitude	Float	Precision geographic coordinate extracted from server token
longitude	Float	Precision geographic coordinate extracted from server token
lead_quality_score	Integer	Balanced calculated score from 0 to 100

---

Configuration

Core Inputs

Field	Type	Required	Description
target_city	String	Yes	City to search, for example "Miami, FL"
target_country	String	No	Restricts the search to a country, for example "US"
search_query	String	Yes	Niche or keyword to search, for example "Dentist"

Spatial Settings

scan_mode controls how the grid spreads across the city.

Value	Behavior
city_center	Keeps the grid tight around the downtown core
balanced	Spreads the grid across main commercial zones
full_coverage	Stretches the grid to the city's outer borders

scan_depth controls how many points are in the grid.

Value	Points	Layout
basic	1	Center only
standard	4	2 x 2
deep	9	3 x 3
enterprise	25	5 x 5

Safety and Filters

Field	Type	Description
max_results	Integer	Stops the run once this many unique leads are saved
max_search_requests	Integer	Hard cap on total lookups (keywords times grid points). Run stops if exceeded
required_fields	Array	Drops leads missing these fields, options are "phone" and "website"

---

Performance

This actor is optimized to run efficiently on standard cloud infrastructure.

• Runs cleanly on 2048 MB of memory allocation
• Locked to single concurrency (max_concurrency=1) to eliminate browser memory leaks and processing spikes
• Cost stays predictable because the budget governor checks your configuration requirements before execution

---

Example Configuration

{
  "target_city": "Miami, FL",
  "search_query": "Dentist",
  "scan_mode": "balanced",
  "scan_depth": "standard",
  "max_results": 2,
  "max_search_requests": 3,
  "required_fields": ["phone", "website"]
}

---

Data Storage on Apify

This actor follows the standard Apify storage model.

Input When a run starts, the configuration object you send is saved as an INPUT record in the run's default Key-Value Store. The scraper reads this record at startup to set the search city, queries, scan settings, and filters.

Output Each validated lead is pushed to the run's default Dataset as it is found. Records are available to query through the Dataset API while the run is still in progress, not only after it finishes.

---

API Usage

Start a Run

curl -X POST "https://api.apify.com/v2/acts/primelabs~google-maps-scraper/runs?token=YOUR_API_TOKEN&memory=2048" \
-H "Content-Type: application/json" \
-d '{
  "target_city": "Miami, FL",
  "target_country": "US",
  "search_query": "Dentist",
  "max_results": 2,
  "max_search_requests": 3,
  "scan_depth": "basic",
  "include_lead_scoring": true,
  "include_row_metadata": false,
  "output_format": "full",
  "required_fields": ["phone", "website"]
}'

Fetch Results

$curl -X GET "https://api.apify.com/v2/datasets/YOUR_DATASET_ID/items?token=YOUR_API_TOKEN"