Deep Email, Phone & Social Media Scraper

Find public business contacts from any list of websites. This actor crawls company sites with HTTP requests only, discovers likely contact pages, extracts emails, phone numbers, social profiles, logos, and business metadata, then returns clean lead records ready for CRM, outreach, enrichment, or research.

No browser. No Playwright. No Puppeteer. Just fast HTTP crawling, Chrome-like requests, smart contact-page discovery, and clean Apify dataset output.

Use it when you have a list of company websites and need answers like:

• What is the best public email for this company?
• Does the website publish phone numbers?
• Which social and messaging channels are linked from the site?
• What is the company name, logo, address, or legal metadata?
• Which page was the contact found on?

---

What this actor extracts

Contacts

• Emails from mailto: links, visible text, HTML, Cloudflare protected emails, JSON-LD, and common [at] / dot obfuscation
• Phone numbers from tel: links, page text, and JSON-LD
• E.164 phone normalization when the number is valid
• Best email, best phone, confidence score, and source page

Social and messaging channels

• Facebook
• Instagram
• X / Twitter
• YouTube
• TikTok
• Pinterest
• Reddit
• Snapchat
• Discord
• Twitch
• GitHub
• Medium
• WhatsApp
• Telegram
• Yelp
• Tripadvisor
• App Store
• Google Play
• Amazon, Etsy, and eBay store/profile links

LinkedIn, Trustpilot, Google Maps, and Threads are intentionally excluded to keep the output focused and avoid duplicating separate scrapers.

Business and brand data

• Company name
• Legal name
• Website title and meta description
• Addresses from structured data when present
• Opening hours from structured data when present
• VAT / tax IDs when visible
• Registration numbers when visible
• Best logo URL
• Favicon
• Apple touch icon
• OpenGraph image
• Twitter card image
• Source evidence for logo/contact extraction when enabled

---

Common use cases

• Lead generation from company website lists
• CRM enrichment
• Agency prospecting
• B2B sales research
• Directory enrichment
• Supplier and vendor research
• Startup, SaaS, ecommerce, and local business contact collection
• Marketing outreach preparation
• Checking whether company websites publish contact details
• Building internal contact intelligence datasets

---

Quick start in Apify Console

1. Open the actor in Apify Console.
2. Paste websites into Website URL(s) or domain(s).
3. For the first test, set Number of websites to process to 10.
4. Keep Pages per website between 3 and 10.
5. Keep Stop early when enough contacts are found enabled to save cost.
6. Run the actor.
7. Open the default dataset and export results as JSON, CSV, Excel, XML, RSS, or HTML.

You can paste full URLs or plain domains. These are all valid:

example.com
https://example.com
https://www.example.com/contact

The actor normalizes plain domains to HTTPS automatically.

---

Recommended input examples

Simple website list

Use this for normal lead enrichment.

{
  "startUrls": [
    { "url": "https://www.cloudflare.com/resource/contact-enterprise-sales/" },
    { "url": "https://www.bluehost.com/contact" },
    { "url": "https://www.wolfssl.com/contact/" }
  ],
  "maxDomains": 10,
  "maxPagesPerDomain": 5,
  "maxDepth": 1,
  "stopWhenFound": true,
  "outputMode": "summary",
  "compactOutput": true
}

Deeper contact discovery

Use this when websites are messy and contacts may be on support, about, legal, team, or office pages.

{
  "startUrls": [
    { "url": "https://example.com" }
  ],
  "maxDomains": 1,
  "maxPagesPerDomain": 15,
  "maxDepth": 2,
  "useSitemap": true,
  "stopWhenFound": false,
  "includeEvidence": true,
  "outputMode": "summary"
}

Import websites from another dataset

Use this when a previous actor produced a dataset with website URLs.

{
  "datasetId": "YOUR_DATASET_ID",
  "maxDomains": 1000,
  "maxPagesPerDomain": 5,
  "outputMode": "summary"
}

The actor reads these fields from input dataset items:

website, url, domain, companyUrl, sourceUrl, finalUrl, startUrl

Evidence and audit mode

Use this when you want to see where each email, phone, social profile, or logo came from.

{
  "startUrls": [
    { "url": "https://example.com" }
  ],
  "includeEvidence": true,
  "outputMode": "summary",
  "compactOutput": true
}

Page-level debugging mode

Use this when you want one dataset row per crawled page.

{
  "startUrls": [
    { "url": "https://example.com" }
  ],
  "outputMode": "pages",
  "includeEvidence": true
}

---

Input fields

Field	Type	Default	Description
startUrls	array	sample URLs	Websites or pages to scan. Use full URLs or plain domains.
datasetId	string	empty	Optional Apify dataset containing website/domain fields.
maxDomains	integer	1	Safety cap for unique websites processed from all inputs. Raise it for real batches.
maxPagesPerDomain	integer	5	Maximum pages fetched per website. Start with 3-10.
maxDepth	integer	1	Link depth. 0 = start page only, 1 = linked contact/about pages, 2 = one more layer.
stopWhenFound	boolean	true	Stops early when a strong email plus phone or socials are found. Saves time and cost.
extractEmails	boolean	true	Extract normal, obfuscated, mailto, JSON-LD, and Cloudflare protected emails.
extractPhones	boolean	true	Extract and validate phone numbers.
extractSocials	boolean	true	Extract social, messaging, marketplace, and app profile links.
extractBrandAssets	boolean	true	Extract logo, favicon, OpenGraph image, and related brand images.
extractBusinessData	boolean	true	Extract company name, legal name, addresses, opening hours, tax IDs, registration numbers.
useSitemap	boolean	true	Reads sitemap.xml and adds contact-like URLs.
followSubdomains	boolean	false	Allows crawling same-root subdomains, for example help.example.com.
countryHint	string	US	Default phone country for numbers without country prefix.
outputMode	string	summary	summary, pages, or both.
includeEvidence	boolean	false	Adds detailed source evidence objects.
compactOutput	boolean	true	Removes empty arrays, empty objects, nulls, and blank strings.
maxConcurrency	integer	10	Number of websites processed in parallel.
requestTimeoutSec	integer	25	HTTP request timeout per page.
maxRetries	integer	3	Retry budget for HTTP and connection errors.
maxProxyRetries	integer	3	Extra retry budget for proxy/transport failures.
proxyConfiguration	object	Apify Proxy off	Apify Proxy settings. Keep it off for cheap tests; enable it if target sites block direct requests.
userAgent	string	empty	Optional custom user agent. Leave empty for built-in Chrome-like headers.

Accepted countryHint values:

US, GB, DE, FR, ES, IT, NL, PL, IN, CA, AU, BR, MX

The API also accepts URL aliases for convenience:

websites, start_urls, urls, domains

---

Output modes

summary

Default mode. Returns one clean lead row per website. Best for exports, CRM import, enrichment, and normal use.

pages

Returns one row per crawled page. Best for debugging, QA, and checking which pages had contacts.

both

Returns summary rows plus page rows. Use only when you need both lead records and page-level evidence in the same dataset.

---

Main output fields

Website identity

Field	Description
recordType	Usually domain in summary mode.
domain	Final website host.
rootDomain	Root domain used for matching.
startUrl	Original normalized input URL.
finalUrl	Final URL after redirects.
status	ok, no_contacts_found, or failed.
statusCode	HTTP status code of the first successful page.
pageTitle	First useful page title.
metaDescription	First useful meta description.
language	HTML language attribute when present.
countryHint	Phone country hint used by the run.
pagesCrawled	Number of pages fetched for the website.
pagesMatched	Number of pages where contacts or socials were found.
crawlDepthReached	Deepest crawl depth reached.

Emails

Field	Description
bestEmail	Best ranked email for outreach.
bestEmailType	sales, support, info, press, jobs, privacy, billing, personal, or unknown.
bestEmailConfidence	high, medium, or low.
emails	Unique email list.
emailDetails	Detailed email evidence when includeEvidence is enabled.

Phones

Field	Description
bestPhone	Best ranked display phone.
bestPhoneE164	Normalized E.164 phone when possible.
bestPhoneConfidence	high, medium, or low.
phones	Unique valid phone list.
phoneDetails	Detailed phone evidence when includeEvidence is enabled.

Social profiles

Field	Description
socialProfiles	Unified list of { platform, url } records.
facebookUrls	Facebook pages/profiles.
instagramUrls	Instagram profiles.
twitterUrls	X / Twitter profiles.
youtubeUrls	YouTube channels or handles.
tiktokUrls	TikTok profiles.
whatsappUrls	WhatsApp links.
telegramUrls	Telegram links.
githubUrls	GitHub organization/user profiles.
mediumUrls	Medium profiles.
Other *Urls fields	Additional supported social, app, or marketplace links.

Brand and company data

Field	Description
companyName	Company name from JSON-LD or site metadata.
legalName	Legal name when structured data provides it.
bestLogoUrl	Best logo/image candidate.
logoUrl	Same as bestLogoUrl, for convenient exports.
logoSource	jsonLd, headerLogo, appleTouchIcon, openGraph, twitterCard, or favicon.
logoConfidence	Confidence of selected logo.
faviconUrl	Favicon URL.
appleTouchIconUrl	Apple touch icon URL.
openGraphImageUrl	OpenGraph image URL.
twitterImageUrl	Twitter card image URL.
brandImages	List of likely brand images.
addresses	Structured addresses when present.
openingHours	Structured opening hours when present.
taxIds	Visible VAT/tax IDs when detected.
vatIds	Alias of taxIds.
registrationNumbers	Visible company registration numbers when detected.

Evidence fields

These appear when includeEvidence is enabled:

Field	Description
bestContactPage	Best page where useful contact data was found.
sourcePages	Crawled pages with counts of found emails, phones, and socials.
contactEvidence	Full contact evidence with value, source URL, page type, confidence, and context.
imageEvidence	Logo and image extraction evidence.
warnings	Non-fatal fetch or parsing warnings.
errors	Fatal run errors for that website.

---

Example output

{
  "recordType": "domain",
  "domain": "wolfssl.com",
  "rootDomain": "wolfssl.com",
  "startUrl": "https://www.wolfssl.com/contact/",
  "finalUrl": "https://www.wolfssl.com/contact/",
  "status": "ok",
  "statusCode": 200,
  "companyName": "wolfSSL",
  "pagesCrawled": 5,
  "pagesMatched": 5,
  "bestEmail": "support@wolfssl.com",
  "bestEmailType": "support",
  "bestEmailConfidence": "high",
  "bestPhone": "+1 (425) 245-8247",
  "bestPhoneE164": "+14252458247",
  "bestPhoneConfidence": "high",
  "emails": [
    "support@wolfssl.com",
    "facts@wolfssl.com",
    "licensing@wolfssl.com"
  ],
  "phones": [
    "+1 (425) 245-8247"
  ],
  "socialProfiles": [
    { "platform": "X / Twitter", "url": "https://twitter.com/wolfssl" },
    { "platform": "Facebook", "url": "https://www.facebook.com/wolfssl" },
    { "platform": "GitHub", "url": "https://www.github.com/wolfssl" }
  ],
  "bestLogoUrl": "https://www.wolfssl.com/wordpress/wp-content/uploads/2020/12/cropped-wolfssl_logo_300px.png"
}

---

Contact discovery logic

The actor does not crawl the whole website blindly. It prioritizes URLs that usually contain contacts:

/contact
/contacts
/about
/team
/staff
/support
/help
/sales
/press
/media
/locations
/offices
/impressum
/imprint
/legal
/privacy

When useSitemap is enabled, it checks sitemap.xml and adds only contact-like URLs from the sitemap. This helps find contact pages that are not linked from the homepage.

---

How ranking works

The actor returns all unique contacts, but it also chooses bestEmail and bestPhone.

Email ranking prefers:

1. Same-domain business emails
2. High-confidence sources like mailto, contact pages, support pages, and JSON-LD
3. Useful role emails like sales, info, support, and press
4. Lower-confidence text matches only when they look legitimate

Phone ranking prefers:

1. Valid phone numbers
2. Numbers matching the selected countryHint
3. Numbers found on contact, support, sales, or legal pages
4. Numbers that can be normalized to E.164

---

Clean output and filtering

The actor filters common junk before writing results:

• Fake emails like test@example.com
• Asset-like strings that look like emails but are actually images or scripts
• Sentry, schema, Wix, and placeholder domains
• Personal/free-mail noise when it appears as unrelated text on a different company site
• Social share/login URLs
• LinkedIn, Trustpilot, Google Maps, and Threads links
• Empty arrays, empty objects, nulls, and blank strings when compactOutput is enabled

This keeps Apify's All fields view clean and makes CSV/Excel exports easier to use.

---

How to run with Apify API

Replace YOUR_TOKEN with your Apify API token.

curl -X POST "https://api.apify.com/v2/acts/trakk~deep-email-phone-social-media-scraper-search/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "startUrls": [
      { "url": "https://www.wolfssl.com/contact/" },
      { "url": "https://www.bluehost.com/contact" }
    ],
    "maxDomains": 2,
    "maxPagesPerDomain": 5,
    "outputMode": "summary"
  }'

To get dataset items after the run finishes:

$curl "https://api.apify.com/v2/datasets/DATASET_ID/items?clean=true&format=json&token=YOUR_TOKEN"

---

Apify CLI commands

Log in

$apify login

Run the actor on Apify

$apify call trakk/deep-email-phone-social-media-scraper-search --input-file input.json

You can also call by actor ID:

$apify call BtjVjAQKexpfdq5po --input-file input.json

Run and print dataset output

$apify call BtjVjAQKexpfdq5po --input-file input.json --output-dataset

Check latest runs

$apify runs ls BtjVjAQKexpfdq5po --limit 10 --desc

View one run

$apify runs info RUN_ID

Download dataset items

apify datasets get-items DATASET_ID --format json
apify datasets get-items DATASET_ID --format csv
apify datasets get-items DATASET_ID --format xlsx

Deploy updates to Apify

$apify push --force

---

Local development commands

Install dependencies:

$pip install -r requirements.txt

Run locally with Apify storage:

$apify run

Run the Python module directly:

$python -m src

Run unit tests:

$python -m unittest discover -s tests -v

Validate the Apify input schema:

$apify validate-schema .actor/input_schema.json

Deploy:

$apify push --force

---

Performance tips

Fast and cheap first run

Use:

{
  "maxPagesPerDomain": 3,
  "maxDepth": 1,
  "stopWhenFound": true,
  "maxConcurrency": 10
}

Better coverage

Use:

{
  "maxPagesPerDomain": 10,
  "maxDepth": 2,
  "stopWhenFound": false,
  "useSitemap": true
}

Large website lists

Recommended settings:

{
  "maxPagesPerDomain": 3,
  "maxDepth": 1,
  "stopWhenFound": true,
  "compactOutput": true,
  "includeEvidence": false,
  "maxConcurrency": 10
}

When to use proxies

Apify Proxy is disabled by default so the sample run stays cheap and stable. For most public company websites, direct requests are enough. If a website blocks direct traffic, enable Apify Proxy in proxyConfiguration; residential proxy can help for more protected sites.

When to raise retries

Raise maxRetries and maxProxyRetries if some websites randomly fail with connection errors, 429, 5xx, or temporary blocks.

---

Status values

Status	Meaning
ok	The website was crawled and at least one contact, phone, or social profile was found.
no_contacts_found	Pages were fetched, but no useful contacts were found.
failed	No pages were crawled successfully. Check warnings or errors.

---

FAQ

Does this actor use a browser?

No. It is requests-only. It does not use Playwright, Puppeteer, Selenium, or a headless browser.

Can it find emails hidden behind JavaScript?

Sometimes, if the email is present in the HTML, JSON-LD, mailto, Cloudflare email protection, or page text. It will not execute JavaScript.

Does it verify that an email inbox exists?

No. It extracts and cleans public emails, but it does not perform SMTP verification or deliverability checks.

Why did a website return no contacts?

Possible reasons: the site blocks automated requests, contacts are loaded only after JavaScript execution, contacts are behind forms, or the site does not publish direct contacts.

Can I scrape thousands of websites?

Yes. Use datasetId for large input lists, keep maxPagesPerDomain modest, keep stopWhenFound enabled, and tune maxConcurrency based on stability.

What is the best setting for normal lead generation?

Use summary output, compactOutput: true, maxPagesPerDomain: 5, maxDepth: 1, stopWhenFound: true, and includeEvidence: false.

When should I enable evidence?

Enable includeEvidence when you need to audit where contacts came from, debug results, or show source URLs to a client. Keep it disabled for cleaner CSV exports.

Can I use the result in Google Sheets, Zapier, Make, or n8n?

Yes. Apify datasets and webhooks work with all common automation tools.

Does it scrape LinkedIn?

No. LinkedIn is intentionally excluded. Use a dedicated LinkedIn actor if you need LinkedIn data.

Is this legal?

The actor extracts publicly visible website data. You are responsible for using the data legally and respecting privacy, anti-spam, GDPR, CCPA, CAN-SPAM, and other rules that apply to your use case.

---

Best practices for clean lead lists

• Start with company homepages or contact pages.
• Keep compactOutput enabled.
• Use summary mode for CRM exports.
• Use includeEvidence only when you need auditability.
• Run a small sample first, then scale.
• For international phone numbers, set countryHint to the most common target country.
• For outreach, verify emails with a deliverability tool before sending campaigns.

---

Tags

email scraper | phone scraper | contact scraper | social media scraper | website contact extractor | lead generation | b2b leads | company enrichment | crm enrichment | business contacts | website scraper | email finder | phone number finder | social profile finder | logo extractor | brand data | sales prospecting | marketing outreach | Apify actor | HTTP scraper | no browser scraper

---

Built for Apify. HTTP-only. Clean contact leads from website lists.