Google Scholar Scraper

Google Scholar Scraper turns Google Scholar into clean JSON, CSV or Excel, without login, cookies or a paid SERP API. Search papers by keyword (with year filters and Scholar operators), read author profiles with h-index, i10-index, total citations and citations per year, and list every paper that cites a given article. Each paper comes with title, authors (with their Scholar profile ids), publication, year, publisher, snippet, PDF link, cited-by count, versions and related links.

US$ 9 per 1,000 results. No monthly fee, no start fee. Error rows are free.

At a glance

• Price per 1,000 results: US$ 9, no run start fee
• Price per author profile: US$ 0.009 (articles charged as results)
• Modes: Keyword search, author profile, cited-by list, or any Scholar link
• Author metrics: Citations, h-index, i10-index (all time and since 2021), citations per year, co-authors
• Paper data: Authors with profile ids, venue, year, PDF link, cited-by count and id, versions, related
• Error rows: Free, with an errorCode

One real row

Search large language models, collected on 2026-09-30 in an Apify cloud run:

{
  "type": "result",
  "source": "search:large language models",
  "position": 1,
  "title": "A survey on evaluation of large language models",
  "url": "https://dl.acm.org/doi/abs/10.1145/3641289",
  "pdfUrl": "https://dl.acm.org/doi/pdf/10.1145/3641289",
  "resultType": "article",
  "authors": [
    { "name": "Y Chang", "profileId": "Hw-lrpAAAAAJ", "profileUrl": "https://scholar.google.com/citations?user=Hw-lrpAAAAAJ&hl=en" },
    { "name": "X Wang", "profileId": "Q7Ieos8AAAAJ", "profileUrl": "https://scholar.google.com/citations?user=Q7Ieos8AAAAJ&hl=en" }
  ],
  "publication": "ACM transactions on …",
  "year": 2024,
  "publisher": "dl.acm.org",
  "snippet": "… 3.1 Natural Language Processing Tasks … the development of language models, particularly large language models, was to enhance performance on natural language processing tasks, …",
  "citedByCount": 6830,
  "citedByUrl": "https://scholar.google.com/scholar?cites=3625720365842685347&as_sdt=2005&sciodt=0,5&hl=en",
  "citesId": "3625720365842685347",
  "versionsCount": 11,
  "totalResults": 6560000
}

(authors list shortened here; the run returned all five linked authors.)

An author profile row collected on 2026-10-21 (DhtAFkwAAAAJ, run with maxArticlesPerAuthor: 0, citations per year shortened):

{
  "type": "author",
  "name": "Kaiming He",
  "affiliation": "Associate Professor, EECS, MIT",
  "verifiedEmailDomain": "mit.edu",
  "interests": ["Computer Vision", "Machine Learning"],
  "citationsAll": 888846,
  "citationsSince": 701747,
  "hIndexAll": 78,
  "hIndexSince": 74,
  "i10IndexAll": 95,
  "i10IndexSince": 94,
  "sinceYear": 2021,
  "citationsPerYear": { "2023": 125281, "2024": 130814, "2025": 132234, "2026": 100851 },
  "articlesCount": 0
}

What you can do with it

• Literature reviews and systematic reviews: every paper for a query and year range, with PDF links and citation counts, ready for a spreadsheet.
• Citation analysis: who cites a paper (citedBy), how citations grow year by year, which versions exist.
• Researcher and lab profiling: h-index, i10-index, citations, interests, affiliation and co-authors for a list of profiles, for hiring, grants, rankings or partnerships.
• Research monitoring: new papers on a topic (sortBy: date) or new citations of your own work.
• Datasets for AI and bibliometrics: titles, snippets, venues and author ids across thousands of papers.

Google Scholar search scraper (papers by keyword)

searchQueries takes anything you would type in Scholar, including "exact phrase", author:"lecun", source:nature and intitle:. Filter with yearFrom/yearTo, sort by relevance or date, and skip [CITATION] entries with excludeCitations. Google Scholar shows up to 1,000 results per query; the Actor pages through them 10 at a time.

Google Scholar author profile scraper (h-index, citations)

authorProfiles takes profile ids (DhtAFkwAAAAJ) or scholar.google.com/citations?user= links. You get one author row plus one result row per article (title, authors, venue, year, cited-by count and link), read 100 per page up to maxArticlesPerAuthor.

Google Scholar "Cited by" scraper

citedBy takes the citesId of any result (the number after cites=) or the whole "Cited by" link, and returns the papers that cite it, with the same fields as a search result.

Input

Field	What it does
searchQueries	Keywords or full Scholar search links
authorProfiles	Profile ids or profile links
citedBy	cites= ids or "Cited by" links
maxResultsPerQuery	Default 50, per query or cited-by list (max 1,000)
maxArticlesPerAuthor	Default 100; 0 returns only the profile row
yearFrom, yearTo	Publication year range (search and cited-by)
sortBy	relevance (default) or date
excludeCitations	Skip [CITATION] entries

{
  "searchQueries": ["crispr gene editing"],
  "authorProfiles": ["DhtAFkwAAAAJ"],
  "citedBy": ["3625720365842685347"],
  "maxResultsPerQuery": 200,
  "yearFrom": 2020
}

Output fields

Results (type: "result"): source, position, title, url, pdfUrl, fullTextUrl, resultType (article, book, citation, pdf, html), authors (name, profileId, profileUrl), authorsLine, publication, year, publisher, snippet, citedByCount, citedByUrl, citesId, clusterId, resultId, versionsCount, versionsUrl, relatedUrl, totalResults. Profile articles also carry citationId, authorProfileId and authorName.

Profiles (type: "author"): profileId, url, name, affiliation, organizationId, verifiedEmailDomain, homepage, interests, photoUrl, citationsAll, citationsSince, hIndexAll, hIndexSince, i10IndexAll, i10IndexSince, sinceYear, citationsPerYear, coAuthors, articlesCount.

Math in titles comes as LaTeX between $ signs, as Scholar stores it (for example $\pi_0$).

Error codes (never charged)

errorCode	Meaning
NO_RESULTS	Google Scholar returned nothing for this search or cited-by id
NOT_FOUND	The profile does not exist or is not public
INVALID_INPUT	A value the Actor does not understand
LOGIN_REQUIRED	The page asks for login (Scholar's author search by name does); use a profile id
BLOCKED	Google Scholar showed a captcha on several IPs in a row; run again
NOT_REACHED, UNEXPECTED	The run timeout arrived, or something unforeseen broke; only that input is lost

Pricing

Event	Per 1,000
Result delivered (search, cited-by or profile article)	US$ 9
Author profile delivered	US$ 9

1,000 papers cost US$ 9. A profile with 100 articles costs US$ 0.909. Duplicates and error rows are free, and there is no Apify compute bill on top.

FAQ

Why not an official Google Scholar API? There is none. Google has never offered a public Google Scholar API, so every tool reads the public pages; this one does it without login and without a paid SERP API in between.

What about captchas? When Scholar shows a captcha, the Actor retries by itself; only if every retry fails does the query become a free BLOCKED row. It never solves or bypasses captchas.

Can I search authors by name? Google Scholar requires login for its author search, so the Actor takes profile ids or links. Search results already include the profile id of every linked author.

Limitations

• Google Scholar caps each query at 1,000 results; split big topics by year range.
• Sorting by date only covers roughly the last year (Scholar's own rule).
• Author names in search results are abbreviated as Scholar shows them (Y Chang); the profile id links to the full profile.
• Only public data. This Actor reads public pages only and respects the site's terms.

Changelog

• 2026-09-30, 0.1: first release. Keyword search with year filters, author profiles with metrics and all articles, cited-by lists, free error rows.