How to Build an Automated B2B Lead Generation Pipeline: Scraping Google Maps, Company Websites & GitHub Contributors with Node.js & Python

[Input URLs: Google Maps / Directories / Company Domains / GitHub Repos] ├──> Layer 1: JSON-LD Parser (LocalBusiness, Organization, AggregateRating) ├──> Layer 2: Same-Domain Subpage Crawler (/contact, /about, /locations) └──> Layer 3: Sanitization & Lead Scoring (Filter telemetry, export CSV/JSON) 

import { CheerioCrawler, createCheerioRouter } from '@crawlee/cheerio'; const EMAIL_REGEX = /[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}/g; const PHONE_REGEX = /(?:\+?\d{1,3}[\s.-]?)?\(?\d{3}\)?[\s.-]?\d{3}[\s.-]?\d{4}/g; const BAD_EMAIL_EXT = /\.(png|jpg|jpeg|gif|svg|webp|ico|css|js|woff|woff2|ttf|map)$/i; const BLOCKED_DOMAINS = ['example.com', 'sentry.io', 'wixpress.com', 'schema.org', 'domain.com']; export function parseJsonLdBusinesses($) { const businesses = []; $('script[type="application/ld+json"]').each((_, el) => { try { const raw = $(el).contents().text().trim(); if (!raw) return; const parsed = JSON.parse(raw); const items = Array.isArray(parsed) ? parsed : (parsed['@graph'] || [parsed]); for (const item of items) { if (item && typeof item === 'object') { businesses.push(item); } } } catch { // Ignore malformed JSON-LD blocks on broken templates } }); return businesses; } export const router = createCheerioRouter(); router.addDefaultHandler(async ({ enqueueLinks, request, $, body, log }) => { const loadedUrl = request.loadedUrl || request.url; const domain = new URL(loadedUrl).hostname.replace(/^www\./, ''); const html = typeof body === 'string' ? body : body.toString('utf-8'); // Automatically follow high-value contact & location pages on the same domain if (!request.userData.isSubPage) { await enqueueLinks({ strategy: 'same-hostname', globs: ['**/contact*', '**/about*', '**/locations*', '**/impressum*'], userData: { isSubPage: true }, }); } const emailSet = new Set(); const phoneSet = new Set(); // 1. Extract explicit mailto: and tel: links $('a[href^="mailto:"]').each((_, el) => { const email = ($(el).attr('href') || '') .replace(/^mailto:/i, '') .split('?')[0] .trim() .toLowerCase(); if (email.includes('@') && !BAD_EMAIL_EXT.test(email)) { emailSet.add(email); } }); $('a[href^="tel:"]').each((_, el) => { const phone = ($(el).attr('href') || '').replace(/^tel:/i, '').trim(); if (phone.length >= 7) phoneSet.add(phone); }); // 2. Scan raw HTML for inline emails with strict false-positive filtering for (const match of html.matchAll(EMAIL_REGEX)) { const candidate = match[0].toLowerCase(); const isBlocked = BLOCKED_DOMAINS.some((d) => candidate.includes(d)); if (!BAD_EMAIL_EXT.test(candidate) && !isBlocked) { emailSet.add(candidate); } } // 3. Scan visible body text for phone numbers (10 to 15 digits) for (const match of ($('body').text() || '').matchAll(PHONE_REGEX)) { const rawPhone = match[0].trim(); const digitCount = rawPhone.replace(/\D/g, '').length; if (digitCount >= 10 && digitCount <= 15) { phoneSet.add(rawPhone); } } // 4. Enrich from Schema.org LocalBusiness / Organization JSON-LD const ldItems = parseJsonLdBusinesses($); let businessName = $('meta[property="og:site_name"]').attr('content') || $('h1').first().text().trim() || $('title').first().text().trim() || domain; let address = $('[itemprop="address"], address').first().text().replace(/\s+/g, ' ').trim() || null; let rating = null; let reviewCount = null; let category = null; for (const ld of ldItems) { if (ld.name && businessName === domain) businessName = ld.name; if (ld.telephone) phoneSet.add(String(ld.telephone)); if (ld.email) emailSet.add(String(ld.email).replace(/^mailto:/i, '').toLowerCase()); if (ld['@type'] && !category) { category = Array.isArray(ld['@type']) ? ld['@type'].join(', ') : String(ld['@type']); } if (ld.address && typeof ld.address === 'object') { const addrParts = [ ld.address.streetAddress, ld.address.addressLocality, ld.address.addressRegion, ld.address.postalCode, ld.address.addressCountry, ].filter(Boolean); if (addrParts.length > 0) address = addrParts.join(', '); } if (ld.aggregateRating && typeof ld.aggregateRating === 'object') { rating = ld.aggregateRating.ratingValue ?? rating; reviewCount = ld.aggregateRating.reviewCount ?? ld.aggregateRating.ratingCount ?? reviewCount; } } // 5. Extract Social Media Profiles const findSocial = (regex) => { let matched = null; $('a[href]').each((_, el) => { const href = $(el).attr('href') || ''; if (!matched && regex.test(href)) matched = href; }); return matched; }; const record = { businessName: String(businessName).replace(/\s+/g, ' ').trim(), category: category || 'LocalBusiness', domain, sourceUrl: loadedUrl, address, rating, reviewCount, emails: Array.from(emailSet), phones: Array.from(phoneSet).slice(0, 10), linkedin: findSocial(/linkedin\.com\/(company|in)\//i), facebook: findSocial(/facebook\.com\/(?!sharer)/i), instagram: findSocial(/instagram\.com\//i), twitter: findSocial(/(twitter|x)\.com\/(?!intent)/i), tiktok: findSocial(/tiktok\.com\/@/i), scrapedAt: new Date().toISOString(), }; log.info(`[Lead Extracted] ${record.businessName} -> ${record.emails.length} emails, ${record.phones.length} phones`); }); 

import requests import pandas as pd APIFY_TOKEN = "YOUR_APIFY_API_TOKEN" ACTOR_URL = ( "https://api.apify.com/v2/acts/" "tungnguyen84~google-maps-lead-email-phone-extractor/" f"run-sync-get-dataset-items?token={APIFY_TOKEN}" ) payload = { "startUrls": [ {"url": "https://apify.com/contact"}, {"url": "https://crawlee.dev"} ], "enrichCompanyWebsites": True, "maxBusinesses": 50 } response = requests.post(ACTOR_URL, json=payload, timeout=180) response.raise_for_status() leads = response.json() df = pd.DataFrame(leads) print(df[["businessName", "domain", "emails", "phones", "linkedin"]]) df.to_csv("enriched_b2b_leads.csv", index=False) 

// Extracting developer lead data from a GitHub profile DOM const username = new URL(loadedUrl).pathname.split('/').filter(Boolean)[0]; const fullName = $('span.p-name, [itemprop="name"]').first().text().trim() || username; const bio = $('div.p-note, [data-bio-text]').first().text().trim() || ''; const company = $('span.p-org, [itemprop="worksFor"]').first().text().trim() || null; const location = $('span.p-label, [itemprop="homeLocation"]').first().text().trim() || null; const website = $('li[itemprop="url"] a, a.Link--primary[rel*="me"]').first().attr('href') || null; const followers = $('a[href$="?tab=followers"] span.text-bold').first().text().trim() || null; 

curl -X POST "https://api.apify.com/v2/acts/tungnguyen84~github-developer-contributor-lead-scraper/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "targetUrls": [ { "url": "https://github.com/tungnguyen84" }, { "url": "https://github.com/apify/crawlee" } ], "maxProfiles": 25 }' 

from collections import defaultdict import json def merge_leads_by_domain(raw_records: list[dict]) -> list[dict]: grouped = {} for rec in raw_records: domain = rec.get("domain") or rec.get("username") or rec.get("sourceUrl") if not domain: continue if domain not in grouped: grouped[domain] = { "businessName": rec.get("businessName") or rec.get("fullName"), "domain": domain, "website"Source：https://dev.to/tungnguyen84/how-to-build-an-automated-b2b-lead-generation-pipeline-scraping-google-maps-company-websites--2kie
My favorites