Table Enrichment with Bright Data

Enrich Excel tables with current web data using Bright Data's 60+ MCP tools.

Requirements

Install openpyxl for Excel file handling:

pip install openpyxl --break-system-packages

Workflow

1. Analyze the table

Read the Excel file and identify:

• Columns that contain identifiers (URLs, usernames, product IDs, company names)
• Columns that need enrichment (empty or to be updated)
• Data type (social media, e-commerce, business, etc.)

import openpyxl

workbook = openpyxl.load_workbook("file.xlsx")
sheet = workbook.active

# Read headers and first few rows to understand structure
headers = [cell.value for cell in sheet[1]]
sample_data = [[cell.value for cell in row] for row in list(sheet.rows)[1:6]]

2. Match tools to data type

Based on identifiers in the table, select appropriate Bright Data tools:

Social Media:

• BrightData:web_data_linkedin_person_profile - LinkedIn profiles
• BrightData:web_data_linkedin_company_profile - LinkedIn companies
• BrightData:web_data_instagram_profiles - Instagram profiles
• BrightData:web_data_tiktok_profiles - TikTok profiles
• BrightData:web_data_youtube_profiles - YouTube channels

E-commerce:

• BrightData:web_data_amazon_product - Amazon products
• BrightData:web_data_walmart_product - Walmart products
• BrightData:web_data_ebay_product - eBay products
• BrightData:web_data_etsy_products - Etsy products

Business:

• BrightData:web_data_crunchbase_company - Company data
• BrightData:web_data_zoominfo_company_profile - ZoomInfo data

General:

• BrightData:scrape_as_markdown - Any public webpage

3. Enrich data row by row

For each row, call the appropriate MCP tool and extract relevant fields:

# Example: Enriching LinkedIn profiles
for row_idx, row in enumerate(sheet.iter_rows(min_row=2), start=2):
    linkedin_url = row[url_column_idx].value
    
    if linkedin_url:
        # Call BrightData:web_data_linkedin_person_profile
        # Extract: name, title, company, location, connections, etc.
        # Write to corresponding columns
        
        sheet.cell(row=row_idx, column=name_col).value = profile_data['name']
        sheet.cell(row=row_idx, column=title_col).value = profile_data['title']

4. Handle errors and missing data

• Skip rows with invalid/missing identifiers
• Log which rows were enriched successfully
• Note rate limits or tool failures
• Preserve original data

5. Save enriched file

Save to /mnt/user-data/outputs/:

output_path = '/mnt/user-data/outputs/enriched_table.xlsx'
workbook.save(output_path)

Provide link: [View enriched table](computer:///mnt/user-data/outputs/enriched_table.xlsx)

Tool selection logic

If table has LinkedIn URLs → Use BrightData:web_data_linkedin_person_profile or BrightData:web_data_linkedin_company_profile

If table has Instagram URLs → Use BrightData:web_data_instagram_profiles

If table has Amazon URLs (with /dp/) → Use BrightData:web_data_amazon_product

If table has product names but no URLs → Use BrightData:search_engine to find URLs first, then scrape

If table has company names → Search for LinkedIn/Crunchbase URLs, then enrich

If URLs are present but platform unknown → Use BrightData:scrape_as_markdown

Best practices

• Batch similar requests: Group rows by data type before calling tools
• Start small: Test enrichment on first 3-5 rows before processing entire table
• Preserve originals: Create new columns for enriched data
• Show progress: Update user after every 10-20 rows processed
• Handle nulls: Explicitly mark cells where data couldn't be retrieved

Example enrichment types

LinkedIn profiles → Add: job title, company, location, connections, industry

Instagram accounts → Add: followers, posts count, engagement rate, bio

Amazon products → Add: price, rating, review count, availability, seller

Company names → Add: LinkedIn URL, website, employee count, funding, industry

Generic URLs → Add: page title, description, key content, last updated