Google Maps is the largest, most up-to-date, and most overlooked source of local business data in B2B. Roughly 200 million active business listings, rich with phone numbers, categories, websites, ratings, and photos, all publicly indexed. Every local B2B outbound campaign we run starts with Google Maps as the primary source for local targets, because no paid database comes close on freshness.
The catch is that Google does not give you a clean export button, and the official API costs add up fast on large pulls. This is the free scraping method we have used to pull tens of thousands of Google Maps records for client campaigns, the filters that make the data usable, and the enrichment steps that turn a raw scrape into an outbound-ready list.
Why Google Maps beats paid local databases
Data freshness is the whole game in local B2B. A restaurant that closed last month still sits in Yelp and Yellow Pages datasets for another six to nine months. Google Maps reflects that closure within days, because Google has direct signal from the business owner, the building, and the search traffic patterns. For outbound, this means fewer bounced emails, fewer disconnected phone numbers, and fewer wasted touches on dead accounts.
Coverage and vertical depth
Google Maps indexes every vertical that has a physical presence. Dentists, lawyers, accountants, home service, fitness, restaurants, retail, construction. For pure B2B service targeting, the local vertical coverage is unmatched. The only categories where Google Maps is weaker are pure online B2B SaaS and agency targeting, because those businesses often do not register a physical presence.
What you get on every record
A standard Google Maps pull returns business name, category, full address, phone number, website, Google rating and review count, hours of operation, and photos. For outreach, the website and phone number are the two fields that matter most. The website gives you a domain for email guessing and decision-maker lookup. The phone number gives you a direct channel for same-day follow-up on any reply.
The free scraping stack
There are three categories of tool that will pull Google Maps for free or near-free. We rotate between them depending on scale.
Browser extensions for small pulls (under 500 records)
Extensions like Instant Data Scraper, Phantombuster's Maps scraper on the free tier, and open-source options like Outscraper's community build handle pulls up to a few hundred records cleanly. Run them in a logged-in Chrome profile, search for your target query on maps.google.com, and trigger the scrape. Data exports to CSV.
Apify actors for mid-scale (500 to 10,000 records)
Apify's Google Maps scraper actor is free to try and runs about 2 dollars per 1,000 records after. For most agency and in-house use, this is the right tier. It handles pagination, multiple search queries in a single run, and exports directly to JSON or CSV. Total cost for a 10,000 record pull runs 15 to 25 dollars, which is below what any paid database would charge.
Self-hosted Puppeteer for scale (10,000 plus records)
At above 10,000 records, run a Puppeteer script on your own infrastructure. Google rate-limits IP addresses after a few hundred requests, so you will need a rotating proxy pool, but this is a one-time engineering investment that pays for itself on the second campaign. We use this tier only for enterprise-wide national pulls.
The query strategy that cuts noise
The single biggest determinant of data quality is how you structure the search query, not the tool. A lazy query returns a huge list of mixed-fit records. A structured query returns a narrower list with almost all targets qualified.
Combine category with geography, not keyword alone
Search 'dental clinics in Austin TX' rather than just 'dental'. Google Maps resolves category plus location far better than keyword searches, and the results come back with consistent classification.
Run separate queries per metro
A single national query for 'dental clinics USA' caps around 100 to 300 results because Google Maps truncates large result sets. Break a national pull into metro-level queries (200 metros covers the US) and aggregate the outputs. This is the difference between 200 records and 40,000 records on the same vertical.
Filter by rating and review count to prioritise live businesses
A dental clinic with zero reviews and no rating is often a closed location or a solo practitioner with no online presence. Filter the final list to records with at least 10 reviews and a rating above 3.5, and you remove roughly 15 to 25 percent of the raw list as low-signal, without losing any real targets.
Records per metro query
80 to 400
Metro queries to cover US
roughly 200
Rating filter floor
3.5+ with 10+ reviews
Cost per 10K records (Apify)
$15 to $25
Turning the scrape into an outbound-ready list
The scrape gives you business name, website, phone number. None of those is a decision-maker email. The enrichment pass is where the list becomes usable for cold outbound.
Step 1. Extract the primary domain
Most Google Maps records return a website URL with utm tags, paths, or subdomains. Strip all of that to get the root domain. This is the input for step two.
Step 2. Pull the owner or primary contact
Run the domain through Apollo or Prospeo company search to find the listed owner, founder, or principal contact. For small local businesses, this is often the same person across all records. For mid-market local operators, it returns a director or manager.
Step 3. Verify and format
Run the final email list through two verifiers plus Scrubby, the same as any other list. Local business email addresses have a higher catch-all rate than SaaS, so the Scrubby step catches more here than on other verticals. Expect 18 to 30 percent of addresses to be catch-alls.
Mistakes that waste Google Maps scrapes
- Running national queries. Always break into metros. A national query returns fewer than 500 records regardless of tool.
- Skipping the rating filter. A low-rating or no-review listing is often closed or inactive.
- Using the phone number without enrichment. The number is the general business line, not the decision maker. Enrich before dialing.
- Pulling the same vertical twice. Tag previous pulls by metro plus category and exclude before the next run.
- Ignoring hours-of-operation data. A business listed as permanently closed is still in the scrape. Filter it out on the final pass.
Google Maps is the fastest, cheapest, freshest local business dataset in the world. The tools to extract it are mature and inexpensive. The only real question is whether you will take 10 minutes to structure the query correctly or spend two days fighting the output. Ten minutes, every time.