ProductHunt Scraper Pro — Extract Listings with Email & LinkedIn Enrichment

Extract structured data from ProductHunt — product names, upvote counts, makers, taglines, website URLs, topics, and more. Optionally crawl each product's website to find email addresses, LinkedIn company pages, Twitter handles, Facebook pages, and phone numbers — all in a single actor run.

Supports four scraping modes: daily/weekly/monthly/yearly leaderboard, keyword search, topic/category browsing, and direct URL scraping.

---

What This Actor Extracts

From ProductHunt (always)

Field	Type	Description
product_name	string	Official product name
tagline	string	One-line product description
description	string	Full product description
upvote_count	integer	Total upvotes received
comment_count	integer	Number of comments
daily_rank	integer	Leaderboard position (#1 = most upvoted)
launch_date	string	Date featured on PH (YYYY-MM-DD)
product_hunt_url	string	Full PH product page URL
website_url	string	Product's own website
topics	array	Category tags (e.g. "AI", "Productivity")
thumbnail_url	string	Product logo/image URL
makers	array	Founders with name, username, Twitter, PH profile
featured	boolean	Whether PH editorially featured this product

From Product Websites (when enrichContacts: true)

Field	Type	Description
emails	array	Email addresses found on the site
linkedin_url	string	LinkedIn company page
twitter_url	string	Twitter/X profile
facebook_url	string	Facebook page
phone_numbers	array	Phone numbers (tel: links)

---

Scraping Modes

1. Leaderboard (default)

Scrape ProductHunt's ranked lists for any time period.

{
    "mode": "leaderboard",
    "leaderboardPeriod": "daily",
    "maxResults": 50
}

Supported periods: daily, weekly, monthly, yearly.

2. Keyword Search

Find products matching a keyword or niche.

{
    "mode": "search",
    "searchQuery": "AI writing tools",
    "maxResults": 100
}

3. Topic / Category

Scrape all products under a ProductHunt topic.

{
    "mode": "topic",
    "topic": "artificial-intelligence",
    "maxResults": 200
}

Popular topic slugs: artificial-intelligence, developer-tools, marketing, productivity, saas, chrome-extensions, no-code, design-tools.

4. Direct URLs

Scrape specific ProductHunt pages.

{
    "mode": "urls",
    "startUrls": [
        { "url": "https://www.producthunt.com/posts/notion-ai" },
        { "url": "https://www.producthunt.com/leaderboard/daily/2026/3/28" }
    ]
}

---

Contact Enrichment Example

Get emails and social links for every product in today's top 50:

{
    "mode": "leaderboard",
    "leaderboardPeriod": "daily",
    "maxResults": 50,
    "enrichContacts": true
}

Output: Two record types in the same dataset, joined on product_hunt_url:

• record_type: "product" — full PH data for each product
• record_type: "contact_enrichment" — emails, LinkedIn, Twitter from their website

---

How It Works

ProductHunt is built on Next.js with Apollo GraphQL. Instead of fragile CSS selector parsing (which breaks every time PH updates their frontend), this actor extracts data directly from the Apollo client cache embedded in window.__NEXT_DATA__. This normalized JSON cache contains all product data in a structured format, making extraction reliable regardless of UI changes.

For contact enrichment, a separate CheerioCrawler visits each product's website and:

• Scans mailto: links for email addresses
• Runs a regex scan of the page text for additional email patterns
• Decodes platform redirect wrappers (/redirect?url=https://linkedin.com/...) before filtering social links — a key detail many scrapers miss
• Extracts LinkedIn, Twitter/X, Facebook, and tel: phone links

---

Pricing

This actor uses PAY_PER_EVENT pricing — you pay only for what you actually use:

• result-found — charged per product record extracted from ProductHunt
• result-enriched — charged per website enrichment record (email/LinkedIn/Twitter)

You can set a budget cap in the Apify console to control total spend per run.

---

Output Format

Results are stored in the default Apify dataset. Filter and join records by record_type:

// Product records
const products = results.filter(r => r.record_type === 'product');

// Enrichment records
const contacts = results.filter(r => r.record_type === 'contact_enrichment');

// Join them
const enriched = products.map(p => ({
    ...p,
    ...contacts.find(c => c.product_hunt_url === p.product_hunt_url),
}));

---

Typical Use Cases

Lead generation for B2B sales — Find newly launched SaaS products and contact their founders before they hit scale.

Market research — Track which categories are growing on ProductHunt week over week.

Competitor monitoring — Get notified when a competitor launches or when products in your category surge in upvotes.

Investor sourcing — Identify high-upvote early-stage products with their maker contact info.

Content creation — Build "Top ProductHunt launches this week" roundups automatically.

AI agent pipelines — Feed ProductHunt data into Claude, ChatGPT, or any LLM agent via the Apify MCP server.

---

FAQ

Does this require a ProductHunt API key? No. This actor scrapes the public ProductHunt website. No API key or account is needed.

How fresh is the data? Data is scraped in real time when you run the actor. Daily leaderboard data reflects the current day's standings at the time of the run.

How accurate is the contact enrichment? Email accuracy depends on how openly each product's website publishes contact info. Most SaaS products publish at least one email. Social links (LinkedIn, Twitter) are found on ~70-80% of product websites.

Can I scrape historical leaderboards? Yes — ProductHunt keeps historical leaderboard pages. Use mode urls and provide the historical leaderboard URL (e.g. https://www.producthunt.com/leaderboard/daily/2025/12/25).

What's the maximum results I can get per run? The maxResults cap is 5,000. For larger datasets, run the actor multiple times with different date ranges or topics.

---

Integration with AI Agents (MCP)

This actor is fully compatible with the Apify MCP server (mcp.apify.com). When connected to Claude, ChatGPT, or Cursor via MCP, AI agents can call this actor directly using natural language:

"Find today's top 20 AI tools on ProductHunt and get their founders' emails"

The agent will set mode: "leaderboard", leaderboardPeriod: "daily", maxResults: 20, enrichContacts: true and return a complete dataset.

---

Support

Found a bug or have a feature request? Leave a review or open an issue on the actor's GitHub page. Pull requests welcome.

Built by: USERNAME/producthunt-scraper-pro