Business Listings Extraction from Google Maps (1 Million Records) – Full Technical Tutorial

Step 6: Extract Business Details (Name, Rating, Category, Location)

records = []

for card in cards:
    try:
        name = card.find_element(By.CLASS_NAME, "qBF1Pd").text
    except:
        name = ""

    try:
        rating = card.find_element(By.CLASS_NAME, "MW4etd").text
    except:
        rating = ""

    try:
        reviews = card.find_element(By.CLASS_NAME, "UY7F9").text
    except:
        reviews = ""

    try:
        category = card.find_element(By.CLASS_NAME, "W4Efsd").text
    except:
        category = ""

    try:
        address = card.find_element(By.CLASS_NAME, "rllt__details").text
    except:
        address = ""

    try:
        url = card.find_element(By.TAG_NAME, "a").get_attribute("href")
    except:
        url = ""

    records.append({
        "name": name,
        "rating": rating,
        "review_count": reviews,
        "category": category,
        "address": address,
        "maps_url": url
    })

Step 7: Extract Place ID (Needed for API-like Calls)

Google Maps URLs contain:

.../data=!3m1!4b1!4m5!3m4!1s!...

Extract Place ID via regex.

import re

def extract_place_id(url):
    match = re.findall(r"1s(.*)!2m", url)
    return match[0] if match else None

for r in records:
    r["place_id"] = extract_place_id(r["maps_url"])

Step 8: Reverse-Engineer Google’s NearbySearch / Details API Call

Google often serves structured JSON via:

https://www.google.com/maps/preview/place/details?authuser=0&hl=en&gl=us&pb=!1m2!1s!...

We can fetch details using Requests:

import requests

def fetch_place_details(place_id):
    try:
        url = f"https://www.google.com/maps/preview/place/details?authuser=0&hl=en&gl=us&pb=!1m2!1s{place_id}"
        r = requests.get(url, headers={"User-Agent": "Mozilla/5.0"})
        return r.text
    except:
        return "{}"

Parse JSON-like structure from Google’s "pb" encoded data:

This requires custom parsers (Actowiz uses internal decoders), but simple fields like phone, hours, plus code, and website can be parsed using regex.

def extract_field(raw, key):
    try:
        idx = raw.index(key)
        snippet = raw[idx:idx+200]
        value = re.findall(r'"([^"]+)"', snippet)[1]
        return value
    except:
        return ""

Attach:

for r in records:
    raw = fetch_place_details(r["place_id"])
    r["phone"] = extract_field(raw, "phone")
    r["website"] = extract_field(raw, "website")
    r["plus_code"] = extract_field(raw, "plus_code")

Step 9: Extract Coordinates (Latitude & Longitude)

Coordinates appear inside listing URLs:

def extract_lat_long(url):
    try:
        match = re.findall(r"@([\d\.\-]+),([\d\.\-]+)", url)[0]
        return float(match[0]), float(match[1])
    except:
        return None, None

for r in records:
    r["latitude"], r["longitude"] = extract_lat_long(r["maps_url"])

Step 10: Handle Scaling (Extracting 1 Million+ Records)

To scrape 1M records:

Strategy 1 — Keyword Expansion

Generate 500+ queries per country:

• “restaurants in <city>”
• “gyms in <city>”
• “pharmacies in <city>”
• “electronics store in <city>”

Strategy 2 — Geo-grid Tiling

Cut the map into grids:

• latitude increments: 0.05
• longitude increments: 0.05

Search each grid:

https://www.google.com/maps/search/restaurants/@,,14z

Strategy 3 — Proxy Rotation

Use:

• residential proxies
• mobile proxies
• IP rotation every 5 requests

Strategy 4 — Parallelization

Run scraper on:

• 50+ Docker containers
• multiple regions
• auto-restart on failure

Strategy 5 — Caching Place IDs

Avoid duplicate businesses across queries.

Step 11: De-duplicate Records

Use Place ID as unique key:

import pandas as pd

df = pd.DataFrame(records)
df = df.drop_duplicates(subset=["place_id"])

Step 12: Export Final Dataset

df.to_csv("google_maps_business_listings.csv", index=False)

Step 13: Build a Lead Gen or Store Locator Dataset

Step 14: Build Analytics From Dataset

• Top business categories
• Density maps
• Heatmaps for store clusters
• Competitor overlap analysis
• Rating distribution
• Review sentiment
• Area-based revenue proxies

Technical Challenges in Google Maps Scraping

Actowiz Solutions solves this using:

• browser fingerprinting
• human-like scroll patterns
• rotating proxies
• load balancing across 30–50 nodes
• place ID caching
• retry-on-failure engines

When Should You Use Actowiz Solutions?

Choose Actowiz if you need:

• 1M–50M Google Maps listings
• Hourly or daily updates
• Multi-country intelligence (USA, UAE, India, Europe)
• Review sentiment analysis
• Category-level heatmaps
• Hyperlocal competitor insights
• API access to Google Maps intelligence
• Fully enriched business listings

We support extraction across:

• Google Maps
• Bing Maps
• Apple Maps
• Yelp
• Zomato
• Swiggy
• TripAdvisor

Conclusion

In this tutorial, you learned how to:

• scrape Google Maps business listings
• scroll dynamically for thousands of entries
• extract place IDs
• fetch details via API-like preview URLs
• extract phone, website, hours
• fetch coordinates
• scale scraping to 1 million+ records
• remove duplicates
• export datasets
• build large-scale business intelligence

This becomes the foundation of a full-scale location intelligence pipeline.

You can also reach us for all your mobile app scraping, data collection, web scraping , and instant data scraper service requirements!