A lead scraper API turns a keyword and a location into contactable business records. This one runs the Google Maps search for you, then visits each business's own website — homepage plus one contact or about page — and collects the emails, phone numbers and social profiles published there.
The two-step design exists because of a fact people discover the hard way: Google Maps does not publish business emails. No Maps scraper returns one, because the data is not on the listing. The email lives on the business's own website, which means someone has to go and read it. That second hop is what separates a directory from a lead list, and it is what you are paying $0.01 per delivered lead for.
$0.01 per delivered lead, up to 500 leads per run. Nothing delivered means nothing charged.
Stage one is a Maps search for your keyword in your location, producing business records with name, category, rating, address, phone and website. Stage two takes each business that published a website and fetches its homepage plus one likely contact page, extracting emails, phone numbers and social profile links.
The crawl is bounded on purpose — two pages per business, not a full site crawl. That keeps the cost per lead predictable and avoids turning a lead run into an unbounded crawl of a thousand small business sites. It also means a contact buried five clicks deep will be missed, which is an honest trade rather than a defect.
only_with_contacts is the input that changes the economics: set it and rows without an email or phone are not delivered, so you are not billed for businesses you cannot contact. site_status and source_pages tell you what happened — whether the site was reachable and which pages the contacts came from, so a suspicious address can be traced back.
query and location are both required — it is a semantic search, not a URL list. only_with_contacts is the input worth setting for most lead work: it suppresses and un-bills rows with no email and no phone. country and lang set the exit and locale for the underlying Maps search. max_results caps delivery and, at a cent per lead, is a real budget control.
One row per business. emails, phones and socials are arrays — a business often publishes several — while phone is the single number Maps listed, which is not always the same as what the website shows. source_pages records which URLs the contacts came from, so anything questionable can be verified rather than trusted. site_status distinguishes a business with no website from one whose website failed to load, which are different findings. place_id and data_id let you chain into the reviews collector for the same businesses.
$0.01 per delivered lead ($10 per 1,000). Nothing delivered means nothing charged, and the $2 monthly free credit covers roughly 200 leads here. Volume tiers take up to 30% off.
$0.01 per delivered lead, $10 per 1,000 — the highest unit price in the catalogue outside company profiles, and for a straightforward reason: each lead is a Maps result plus two website fetches, so it is three fetches wearing one price.
A run of 500 dentists in a metro area is $5. With only_with_contacts enabled you pay only for the ones you can actually contact, which on most categories is a substantial share of the total. The free $2 monthly allowance covers about 200 leads.
The commercial contact databases are the real alternative here, and they are good at something this is not: mapping people to companies. If you need the marketing director's direct dial at a mid-size firm, that is what they sell, and no amount of website scraping substitutes for it.
The trade-offs run the other way for local business data. Those databases skew heavily toward companies with a professional-network footprint, which means small local businesses — the dentist, the plumber, the independent restaurant — are thinly covered or absent. Their records also age: a contact database is a snapshot of when the record was last verified, whereas a website read today shows the address the business is publishing today. And they price by seat and annual contract rather than per row.
The honest split: named decision-makers at larger companies, buy a database. Contactable local businesses in a category and city, collect them.
Teams usually build this in two weekends and maintain it for years. The parts that cost more than expected:
name [at] domain [dot] com. Naive regex misses most of it and hallucinates some.If you only need the business list without contacts, the Maps collector at $0.001 is ten times cheaper — use it when you do not need the enrichment.
Every business in a category and city with an email or phone, ready for a sequenced outreach campaign — subject to the marketing law in your market.
Size an addressable market before committing to it: how many businesses, how many contactable, how many with a working website at all.
site_status surfaces businesses whose site is broken or missing — a qualified list for anyone selling web services.
Match on name and address, then attach the emails and social profiles the business publishes itself.
Up to 500 leads per run, which is lower than the search collectors because each lead involves three fetches. Website enrichment is bounded at homepage plus one contact page — a deeply buried contact will be missed. Businesses with no website return Maps data only, and only_with_contacts filters them out if you would rather not pay for them.
The legal position deserves real attention here, because collection and use are governed separately. Collecting publicly published business contacts is generally lawful in most jurisdictions. Using them for unsolicited marketing is governed by GDPR and PECR in Europe, CAN-SPAM in the US, CASL in Canada and their equivalents — and some of those, notably CASL and several EU implementations, require consent before you send anything. A sole trader's email is personal data even when it is on a business website. Get advice on the outreach, not just the collection; this is not it.
Not from Maps itself — Google does not publish business emails on a listing, so no Maps scraper can return one. This collector gets them by doing the second step: it visits each business's own website, homepage plus one contact page, and extracts what is published there.
There is a free allowance of $2 a month with no card, about 200 delivered leads at $0.01 each — enough to test a category and a city before committing. Free lead scrapers on GitHub generally do the Maps half and leave the website enrichment, which is the expensive half, to you.
$0.01 per delivered lead — $10 per 1,000. It is the most expensive unit outside company profiles because each lead is a Maps result plus two website fetches. Setting only_with_contacts means you pay only for businesses you can actually contact.
You still get the Maps record — name, category, rating, address and the phone number Google lists — with site_status explaining that there was no site to enrich. Enable only_with_contacts to suppress and un-bill rows with neither an email nor a phone.
They are extracted from pages the business publishes, and source_pages records exactly which URL each one came from, so anything questionable can be verified. The bounded crawl — homepage plus one contact page — means a contact buried deep in a site will be missed rather than guessed at.
That depends on where you and the recipient are, and it is a separate question from whether collection was lawful. GDPR and PECR in Europe, CAN-SPAM in the US and CASL in Canada all govern unsolicited commercial messages, and some require consent before the first send. A sole trader's address is personal data even on a business site. Take proper advice before running a campaign.
The Maps collector returns the business listing at $0.001 per place — name, phone, website, rating, coordinates. This one adds the website enrichment step for emails, additional phones and social profiles, at $0.01. If you do not need contacts beyond the listed phone, the cheaper one is the right call.
Skip the marketing. QuantumProxies.io publishes a machine-readable site map, Markdown for every page, a free MCP server, and APIs billed from the same balance as your proxies.
Connect in one command: npx -y quantumproxies-mcp