How to Extract Business Emails and Social Profiles at Scale with the Website Scraper API

Why Scraping Contact Pages at Scale is Hard

Modern business websites are a contact extraction nightmare. The same company may protect its email behind Cloudflare's email obfuscation (data-cfemail), render its contact form dynamically via JavaScript, bury phone numbers in image-alt text, and list its LinkedIn company page only on a hidden /about-us subpage.

This is precisely why teams building B2B prospecting data at scale don't write custom scrapers for each site — they use a purpose-built API that handles all of these edge cases automatically. The Bolt Scraper Website Scraper API accepts up to 10,000 URLs per job and returns a structured JSON record for each domain, including filtered emails, phones, and extracted social profile links.

Sending Bulk URLs in a Single POST Request

The core endpoint accepts an array of URLs and immediately queues them for parallel cloud extraction. You do not need to loop through domains one by one — the entire list ships in one call, and the scraper handles concurrency for you.

POST https://boltscraper.com/api/v1/website/jobs
Authorization: Bearer bs_YOUR_API_KEY
Content-Type: application/json

{
  "urls": [
    "https://example-co.com",
    "https://demo-roofing.com",
    "https://sunsetshingles.net",
    // ... up to 10,000 URLs
  ]
}

The API responds immediately with a job_id. Scraping runs asynchronously in the cloud. You can either poll for completion or — far more efficiently — register a webhook that fires as soon as your batch finishes.

{
  "success": true,
  "job_id": "job_a3f8c2...",
  "job_name": "Q4 Outreach — Roofing Companies",
  "total_urls": 2847,
  "status": "scraping"
}

Deep Subpage Crawling — Finding /contact, /about, and /team

For every URL submitted, the scraper performs a multi-stage crawl. It starts on the root page and then automatically discovers and follows the contact, about, and team subpages for that domain. This is where 60–80% of direct sales contacts live.

The resulting record for each URL contains up to three deduplicated emails and two phone numbers, plus all discovered social links, already classified and ready to use:

{
  "url":             "https://example-co.com",
  "page_title":       "Example Co",
  "meta_description": "Expert roofing services in Phoenix, AZ",

  // Up to 3 deduplicated emails
  "email_1": "[email protected]",
  "email_2": "[email protected]",
  "email_3": "",

  // Up to 2 phone numbers
  "phone_1": "+15550100192",
  "phone_2": "",

  // Social media — direct profile URLs
  "linkedin_url":   "https://linkedin.com/company/acme-roofers",
  "facebook_url":   "https://facebook.com/example-co",
  "instagram_url":  "https://instagram.com/example_co",
  "twitter_url":    "",
  "youtube_url":    "",

  "status":      "success",
  "scraped_at":  "2026-09-19T11:30:00.000Z"
}

Smart Email Filtering — Direct Contacts Only

Raw scrapers return hundreds of junk email addresses per domain: server notification addresses, CMS system addresses, placeholder addresses from contact form plugins. The Bolt Scraper engine applies a multi-layer filtering stack automatically before returning any email.

This means every email in your results file is a real, deliverable business address that was deliberately placed on the site. Your cold outreach bounce rate drops to near zero before you even hit a verification service.

What Social Platforms Are Extracted

Beyond emails, the scraper parses every outgoing link in the page HTML and classifies social profile URLs by platform. All 12 major platforms are detected and returned as separate fields in your result:

End-to-End Example: Enriching 500 Domains

Here's a complete Python example that submits 500 domains, polls for completion, and then saves the enriched contacts to a CSV:

import requests, time, csv

API_KEY  = "bs_YOUR_API_KEY"
BASE_URL = "https://boltscraper.com/api/v1/website"
HEADERS  = {"Authorization": f"Bearer {API_KEY}"}

# 1 — Submit the batch
domains = ["example-co.com", "demo-roofing.com", ...]  # your domain list

r = requests.post(f"{BASE_URL}/jobs", headers=HEADERS, json={
    "job_name": "Q4 Roofing Enrichment",
    "urls": domains
})
job_id = r.json()["job_id"]
print(f"Job created: {job_id}")

# 2 — Poll until complete
while True:
    status = requests.get(f"{BASE_URL}/jobs/{job_id}", headers=HEADERS).json()
    job = status["job"]
    print(f"  {job['processed_count']}/{job['total_urls']} processed")
    if job["status"] == "COMPLETED":
        break
    time.sleep(15)

# 3 — Fetch results and save to CSV
results_csv = requests.get(
    f"{BASE_URL}/jobs/{job_id}/download",
    headers=HEADERS
)

with open("enriched_contacts.csv", "wb") as f:
    f.write(results_csv.content)

print(f"✅ Results saved to enriched_contacts.csv")

Frequently Asked Questions