Google Search to Full Article Text ⚡$4 per 1k

Search Google and extract full article text from result URLs.

✨ Highlights

• 💸 Fixed price: $4 per 1,000 extracted articles
• ✅ Pay per result: you are billed only when the Actor successfully extracts article text
• 📍 Precise search location: set where the Google search is executed, down to specific cities and regions
• 🗓️ Time range: search within the last N days
• 🔎 Supports Web and News search modes
• 🔑 Multiple keywords per run (up to 10)

🧠 How it works

1. Runs Google search for each keyword using your settings
2. Collects result URLs
3. Opens each URL and extracts the main article text
4. Saves output items into the Apify dataset

🧾 Input format

Provide JSON with these fields:

• queries (array, required): list of search queries, max 10

• articles_limit (integer, required, default: 100): results per keyword (10 to 1000)

• days_back (integer, optional): lookback window in days

• geo_code (integer, optional): search location code (city or region), use the numeric code from the reference list in the input UI

• domain (string, optional): Google domain zone (example: "com", "co.uk", "de", "fr")

• tbm (string, optional): search type

  • "" for web
  • "news" for news

• device (string, optional): "desktop" or "mobile"

• safe (integer, optional): safe search mode, 1 filter, 2 off

Example input:

{
  "queries": ["artificial intelligence", "machine learning"],
  "articles_limit": 50,
  "days_back": 7,
  "geo_code": 1234,
  "domain": "com",
  "tbm": "news",
  "device": "desktop"
}

📦 Output (Apify dataset)

Each item typically contains:

• title: article title
• link: article URL
• content: extracted main text
• query: the search query that produced the result
• pubDate: publication date (if detected)
• additional fields when available (position, source domain, etc.)

Example:

{
  "title": "Example article title",
  "link": "https://example.com/article",
  "content": "Full extracted article text...",
  "pubDate": "2025-10-12",
  "query": "AI chips"
}

📝 Notes

🧱 Access gate text instead of an article: some websites respond with a blocking notice (anti bot page, cookie wall, login prompt, subscribe page, access denied). In some cases, that notice may be returned as plain text and can be mistakenly treated as article content. We filter common patterns, but we do not guarantee 100% detection because we avoid expensive per page AI validation to keep the price low.

🔒 Protected pages: sometimes content is returned by the website even though the publisher would prefer to restrict it (for example partial paywalls, inconsistent access rules, or temporary misconfigurations). If the page returns readable article text at request time, the Actor may extract it.

📅 Publication dates are best effort: many pages do not expose a reliable pubDate.

⚖️ Legal

This Actor is a general purpose text extraction tool. It uses standard HTTP requests and parses only what a target website returns at run time. It is not designed to bypass login systems, paywalls, encryption, DRM, or other access control measures, and it does not include features intended for circumvention.

If a website returns full text without requiring credentials or special access, the Actor may extract it as part of normal processing. We do not claim ownership of third party content and we do not grant any license to republish it. If you are a rightsholder and believe your content is being processed in a way you do not want, contact us and we will help with a prompt removal or blocking approach.

📩 Support and feature requests

Want a specific parameter, output field, or workflow integration? Email: ohmycode.lab@gmail.com