Complete Restaurant Directory Data Scraping: iens.nl & Eet.nu

About the Client

Location: Amsterdam, Netherlands

Industry: Market Research / Food & Beverage Analytics

Objective: To build a comprehensive dataset of all restaurants listed on two major Dutch directory sites—iens.nl and eet.nu—capturing:

• Restaurant name
• Business e-mail address
• Total number of reviews
• City where located

The client required no business to be missed, with results delivered in Excel (.xlsx/.csv) format as quickly as possible.

Actowiz Solutions was engaged because of its experience in directory scraping, data enrichment, and full-scale extraction pipelines.

The Challenges

Scraping large directories like iens.nl and Eet.nu involves several complexities:

• Large Volume & Pagination: Each site lists thousands of restaurants, across multiple pages, filters, and regions. You must crawl all pages without missing any entry.
• Dynamic Loading & JavaScript: Some content is loaded dynamically (AJAX), requiring headless browsers or script execution to uncover all entries.
• Email Harvesting Difficulty: Many listings do not directly show email addresses. Some require visiting the restaurant's official site or contact page to extract the address.
• Data Inconsistencies: Restaurant names, cities, and review counts may be formatted differently across listings, making normalization necessary.
• Rate Limits & Blocking Risk: Aggressive crawling may trigger defenses or temporary bans. The scraper must include delays, proxy rotation, and error recovery.
• Completeness Assurance: Ensuring that no restaurant is left out demands robust link discovery and sanity checks against directory totals.

Project Goals

Actowiz Solutions committed to:

• Extracting every restaurant listing from iens.nl and Eet.nu
• Capturing the four key data points per restaurant
• Enriching with email addresses wherever possible
• Delivering a clean Excel (XLSX/CSV) with distinct columns
• Ensuring high data completeness and accuracy
• Providing data quickly, with progress milestones

Architecture & Workflow

[ iens.nl & Eet.nu Directory Pages ]

↓ [ URL Discovery & Pagination Crawler ]

↓ [ Listing Page Scraper (extract name, city, review count, email link if available) ]

↓ [ Restaurant Site Visit (if needed) to find email ]

↓ [ Data Cleaning & Normalization ]

↓ [ Deduplication & Completeness Checks ]

↓ [ Export to Excel / CSV ]

Key Steps:

Pagination & Region Discovery

Crawl all region and city-level pages to find all restaurant listing URLs.

Listing Extraction

On each listing page, extract:

• Restaurant name
• City
• Total reviews
• Email (if present)

Fallback Email Discovery

If no email on listing, follow the restaurant's website link (if available) and visit "Contact" or "About" pages to find email via regex patterns.

Cleaning & Normalization

• Standardize city names
• Convert review counts to integer
• Remove whitespace and encoding errors

Deduplication & Completeness Assurance

• Use restaurant name + city combination to detect duplicates
• Cross-check listing counts per region with totals reported on directory (if available)
• Flag missing entries for audit

Export Formats & Delivery

Output Excel file with columns: Business Name, Email, Total Reviews, City

Also deliver CSV for compatibility

Performance & Milestones

Total Listings Extracted (Example): 9,500+

Email Addresses Found: ~6,800 (72%)

Cities Covered: ~220

Project Duration: ~6 days

Completeness Estimate: >98% listing coverage

Data Accuracy: ~99% on name/ city / review counts

Chart Idea – Email Coverage vs Listings by City

Create a bar chart per city (top 10 cities) showing:

• Total restaurant listings
• Number of those with email extracted

This highlights which cities had higher email coverage and where fallback discovery was needed.

Infographic Layout – "How We Built a Full Directory Scraper"

Implementation Highlights

• Playwright or headless browser used for portions of listings or dynamic loading.
• Retry and fallback logic for pages failing due to timeouts or changes.
• Proxy rotation and crawl delays to minimize blocking risk.
• Parallel crawling (within safe concurrency limits) to speed up extraction.
• Logging module captures each failed URL, reason, retry count, and timestamp.

Example log entry:

Results & Insights

• High Email Capture Rate: ~72% of listings had an email directly or via fallback.
• City Distributions: Amsterdam had ~1,200 listings, Rotterdam ~800, Utrecht ~550, etc.
• Top Cities by Email Coverage: Smaller towns had higher rates (fewer "no email" cases).
• Review Counts Variability: Some listings had zero reviews; others had 500+ in major cities.

These insights help downstream analytics—for instance, targeting restaurants with reviews >100 and valid email.

Data Quality & Limitations

Known Gaps:

• Some restaurants may use contact forms only (no email in text), which cannot be reliably extracted.
• A few listings had email obfuscation (e.g. 'contact [at] domain'), requiring manual decoding.
• Some dynamic listings only fully load via infinite scroll; small number of edge-case listings may get missed.

Mitigation Approach:

• Flag entries with missing email for manual review.
• Use heuristics and pattern matching to guess obfuscated emails (e.g. replace [at] tokens).
• Cross-validate counts of listings per region with directory metrics (if available).

Impact for the Client

• Delivered a complete, email-rich directory of restaurants across both platforms.
• Replaced weeks of manual data collection with automated pipelines.
• Enabled targeted outreach campaigns (email marketing, partnerships).
• Provided dataset in ready-to-use Excel format—integrated with the client's CRM.

Why Actowiz Solutions

• Proven expertise in large-scale directory scraping and email enrichment.
• Experience with dynamic site rendering, fallback logic, and robust logging.
• Ability to deliver clean, verified output on tight timelines.
• Focus on compliance, error handling, and completeness.

Next Steps / Enhancements

• Expand to other directories (TripAdvisor NL, Yelp NL, etc.).
• Add additional fields like phone numbers, website URLs, opening hours, cuisine type.
• Automate periodic updates (weekly/monthly) to keep data fresh.
• Build a dashboard to visualize email coverage by city, review distribution, etc.

Conclusion

This case study shows how Actowiz Solutions tackled full-scale directory scraping across two large restaurant listing platforms in the Netherlands. The project delivered:

• Nearly complete listing coverage
• High email extraction rate
• Clean, normalized records
• Rapid turnaround

With robust retry logic, fallback paths, and classification, the client now has a marketing-ready dataset for outreach and analytics.