Google Scholar Scraper

Google Scholar Scraper — Papers & Citations

Google Scholar Scraper exports Google Scholar search results as structured data. Each paper comes with its title, authors, venue and year, publisher, abstract snippet, "Cited by" count, number of versions, paper and PDF links, and the links to its citing and related articles. Search by keyword or by author name, filter by year range, and request up to 500 papers per query for USD 0.007 per paper on the FREE tier (USD 0.005 on GOLD). No Google account, cookies or API key are needed; the Actor handles data access.

What you get

• One record per unique result. Results are deduplicated by Google Scholar's cluster ID across all queries in a run, and each one is charged once.
• Citation metrics. citationCount, versionsCount, plus citationsUrl and relatedUrl for citation-graph work.
• Clean bibliographic fields. authors as a list, publicationInfo (venue and year), publisher, and year as a number.
• Full-text links. pdfUrl whenever Google Scholar shows a PDF or HTML full-text link.
• Filters that Google Scholar applies. Year range, sort by relevance or date, and exclusion of citation-only entries.
• Honest run status. Every run writes an OUTPUT record: DATA, PARTIAL, EMPTY, FAILED, REJECTED or CONFIGURATION, with a stop reason per query.

Input

Field	Default	Description
searchQueries	["urban heat island"]	Keywords or topics; up to 20 queries and author names per run
authorUrls	[]	Author names, searched with Google Scholar's author: operator (profile URLs are not supported)
maxResults	20	Papers per query or author, 1–500 (10 per Google Scholar page)
yearLow / yearHigh	none	Publication year range
sortBy	relevance	relevance or date (Google Scholar's date sort covers papers added in the last year)
articlesOnly	true	Skip citation-only entries without a linked document

{ "searchQueries": ["graph neural networks"], "authorUrls": ["Yann LeCun"], "maxResults": 100, "yearLow": 2020 }

Output

{
  "resultId": "tE5liJRnkYMJ",
  "rank": 1,
  "title": "A review on the generation, determination and mitigation of Urban Heat Island",
  "authors": ["AM Rizwan", "LYC Dennis"],
  "publicationInfo": "Journal of environmental sciences, 2008",
  "publisher": "Elsevier",
  "abstract": "… temperature or heat content called Urban Heat Island (UHI) …",
  "citationCount": 2949,
  "versionsCount": 8,
  "paperUrl": "https://www.sciencedirect.com/science/article/pii/S1001074208600194",
  "pdfUrl": "http://www.upv.es/contenidos/CAMUNISO/info/U0564937.pdf",
  "sourceType": "PDF",
  "year": 2008,
  "citationsUrl": "https://scholar.google.com/scholar?cites=9480472578232766132&as_sdt=2005&sciodt=0,5&hl=en",
  "relatedUrl": "https://scholar.google.com/scholar?q=related:tE5liJRnkYMJ:scholar.google.com/&scioq=urban+heat+island&hl=en&as_sdt=0,5",
  "scrapedAt": "2026-10-30T12:00:00.000Z",
  "searchQuery": "urban heat island"
}

Google Scholar truncates long author lists, so authors holds the names it shows. Export the dataset as JSON, CSV, Excel or XML, or read it through the Apify API.

Pricing

Pay per event, with Apify platform usage included.

Event	FREE	BRONZE	SILVER	GOLD+	When it is charged
Paper (apify-default-dataset-item)	$0.007	$0.007	$0.007	$0.005	Once per unique paper saved to the dataset
Actor start (apify-actor-start)	$0.005	$0.005	$0.005	$0.005	Once per run at the default 256 MB

Example: 20 papers cost about $0.15 on FREE ($0.005 + 20 × $0.007) and $0.11 on GOLD. A query with no matches costs nothing beyond the start event.

How it works and limits

• Each Google Scholar page holds up to 10 results and takes about 10–20 seconds to load; a 20-paper query usually finishes in under a minute.
• A page that cannot be loaded is retried; if it still fails, the run keeps every paper already saved and reports the query as PARTIAL.
• The run stops at maxResults, at the last result page, at your maximum charge, or 45 seconds before the run timeout.
• Google Scholar limits how deep automated access can page. In tests on 2026-10-30, broad queries returned 19–33 papers over the first 3–5 pages, then Scholar stopped serving further pages. The run then reports the query as PARTIAL, shows Scholar's total-results estimate, and charges only for papers delivered. For more coverage, split a topic into narrower queries or year ranges.
• Google Scholar itself shows at most about 1,000 results per search.

API example

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('muhammadafzal/google-scholar-scraper').call({
  searchQueries: ['transformer interpretability'],
  maxResults: 50,
  yearLow: 2022,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(`${items.length} papers`);

The Actor can also be called by AI agents through the Apify MCP server.

FAQ

How do I export Google Scholar results to CSV or Excel?

Run the Actor with your keywords, then export the dataset from the run page as CSV or Excel. Each row is one paper with its authors, year, venue and citation count.

Can I get citation counts from Google Scholar without an API?

Yes. Every record has citationCount, the "Cited by" number Google Scholar shows, plus citationsUrl, the list of citing papers.

How do I find all papers by an author?

Put the author's name in authorUrls, for example "Geoffrey Hinton". The Actor searches with Google Scholar's author: operator. Common names can match other researchers, so add a keyword query to narrow it.

How many papers can I get per run?

You can request up to 500 per query or author and 20 queries per run, on every Apify plan including the free plan. Google Scholar usually stops serving pages after the first 20–40 results of a query, so narrower queries or year ranges return more papers in total.

What happens when nothing matches?

The query is reported as EMPTY with the stop reason "no matching articles on Google Scholar", and no paper events are charged.

Related Apify Actors

• 📰 Extract Google News Articles — AI & RAG Ready
• GitHub Repo Search — Stars, Language & Topics
• Email Validator — Syntax, MX & Risk Checks

Last verified: 2026-10-30