Step 1: Starting with Website Input

The workflow begins with a Local Upload component (Node 0) where you provide the target website URL. This serves as the starting point for the entire scraping process, telling the system which website to analyze and extract information from.

Step 2: Homepage Scraping

Once the target website is identified, the Homepage Scraper (Node 1) goes to work. This component performs a comprehensive analysis of the website’s homepage, extracting several key elements:

1. Website title
2. Meta description
3. Main headings (H1 tags)
4. Primary content text
5. Internal links (pages within the same domain)
6. External links (links to other websites)

This initial scraping provides a foundation of information about the website and maps out its structure, which will be crucial for further data extraction.

Step 3: Language Detection

With the basic website information collected, the Detect Language component (Node 2) analyzes the title, meta description, and H1 headings to determine the primary language of the website. This information helps in appropriate processing of the content and ensures accurate extraction of contact details regardless of the language used.

Step 4: Finding the Contact Page

One of the most important steps in the workflow is identifying the Contact URL (Node 3). This component scans through the internal links gathered in Step 2 to find the most likely contact or “about us” page. It prioritizes pages with keywords like “contact,” “imprint,” or “impressum” (in German websites), which typically contain the richest source of contact information.

Step 5: Social Media Profile Identification

Simultaneously, the Social URL Grabber (Node 4) analyzes the external links to identify and extract links to social media profiles. This component looks for URLs that point to platforms like Facebook, LinkedIn, Instagram, Twitter/X, and TikTok, which can provide additional points of contact and information about the organization.

Step 6: Website Categorization and Summary

To provide context for the contact information, the workflow includes components for Website Summary (Node 5) and Website Category (Node 11). These analyze the website content to generate a concise summary of what the organization does and categorize it with appropriate tags, giving you valuable insights about the entity you’re contacting.

Step 7: Social Media Profile Extraction

The workflow includes dedicated components (Nodes 6-10) for extracting specific social media profile URLs. Each platform has its own extraction node:

1. Facebook URL Extractor (Node 6)
2. LinkedIn URL Extractor (Node 7)
3. Instagram URL Extractor (Node 8)
4. Twitter/X URL Extractor (Node 9)
5. TikTok URL Extractor (Node 10)

These components filter and clean the social media URLs to ensure you get direct links to the organization’s profiles on each platform.

Step 8: Contact Page Content Extraction

With the contact page identified in Step 4, the Contact Scraper (Node 12) visits this page and extracts all of its content. This typically includes the most valuable contact information, such as phone numbers, email addresses, physical addresses, and sometimes even specific contact persons.

Step 9: Comprehensive Contact Data Compilation

The Contact Data Extraction component (Node 13) serves as an aggregator for all the contact information found throughout the website. It compiles data from both the homepage and the dedicated contact page, with priority given to information from the contact page as it’s typically more accurate and up-to-date.

This component makes intelligent decisions about how to handle multiple instances of the same type of information (like several phone numbers or addresses), generally prioritizing the first or most prominent instance.

Step 10: Specific Contact Detail Extraction

The final stage of the workflow consists of specialized components (Nodes 14-22) that extract and format specific types of contact information:

1. Company Name Extractor (Node 14)
2. Phone Number Extractor (Node 15)
3. Email Extractor (Node 16)
4. Address Extractor (Node 17)
5. City Extractor (Node 18)
6. Country Extractor (Node 19)
7. Postal Code Extractor (Node 20)
8. First Name Extractor (Node 21)
9. Last Name Extractor (Node 22)

Each of these components uses advanced pattern recognition and natural language processing to identify and extract its specific type of contact information from the aggregated data, ensuring accuracy and completeness.