Medium Scraper

Medium Scraper turns public Medium articles into clean JSON, CSV or Excel, without login or cookies. Search any keyword, read tag feeds (latest, hot, top this week, month, year or all time), pull the archive of an author or a publication, or paste article links. Every article comes with title, subtitle, author and follower count, publication, date, claps, responses, reading time, word count, tags and the member-only flag, plus the full text in plain text and Markdown when you want it.

US$ 2.00 per 1,000 articles, plus US$ 0.80 per 1,000 when you add the full text. No monthly fee, no start fee. Error rows are free.

At a glance

• Price per 1,000 articles: US$ 2.00 (US$ 2.80 with full text; down to US$ 1.70 on Gold and above)
• Inputs: Keyword search, tags x 6 sort orders, authors, publications, article links
• Full text: Plain text and Markdown; member-only articles get Medium's public preview, flagged
• Date filter: publishedAfter as a date or "last N days"
• Login, cookies: None
• Error rows: Free, with an errorCode

One real row

Keyword rust async, full text on (collected on 2026-09-30, text trimmed):

{
  "type": "article",
  "id": "43e4ecf62d2e",
  "title": "Rust Async/Await Explained What Really Happens Behind the Scenes",
  "url": "https://medium.com/@highsongs2020/rust-async-await-explained-what-really-happens-behind-the-scenes-43e4ecf62d2e",
  "publishedAt": "2026-09-22T07:32:01Z",
  "claps": 2,
  "responses": 0,
  "readingTimeMinutes": 4.2,
  "wordCount": 1070,
  "isMemberOnly": true,
  "tags": ["rust", "programming", "software-development", "software-engineering", "backend-development"],
  "authorName": "SOVANNARO",
  "authorFollowers": 52,
  "contentMarkdown": "## Rust Async/Await Explained What Really Happens Behind the Scenes\n\n### Go beyond writing .await ...",
  "contentIsPreview": true,
  "source": "search:rust async",
  "rank": 1
}

What you can do with it

• Content research and SEO: see what gets claps and responses in your topic this week, month or year, and which angles are already covered.
• Author and publication tracking: schedule a run on the writers and publications you follow and get every new article as a row.
• Datasets for AI: full text in Markdown, ready for RAG, summarization or training pipelines (open articles in full, member-only as preview).
• Market signals: track how often a product, library or company shows up in Medium articles over time.

Input

Field	What it does
searchQueries	Keywords. Medium's relevance order, 10 per page, up to maxItems.
tags	Tag slugs or links. Up to 25 articles per sort order.
tagSort	latest, hot, top_week, top_month, top_year, top_all_time (several allowed).
authors	@username, profile link or name.medium.com. Paged archive.
publications	Slug, custom domain or link. Paged archive.
articleUrls	Article links or 12-character ids.
maxItems	Max articles per keyword, tag sort, author or publication. Default 50.
publishedAfter	YYYY-MM-DD or a number of days.
includeContent	Add content and contentMarkdown.
excludeMemberOnly	Skip articles behind the paywall.
onlyNew	Only articles not delivered in earlier runs, for schedules. Default off.

{ "tags": ["startup", "product-management"], "tagSort": ["top_week", "top_month"], "maxItems": 25 }

{ "publications": ["pub.towardsai.net"], "publishedAfter": "30", "includeContent": true, "maxItems": 200 }

Schedule it daily: only new articles

Turn on onlyNew, save the input as a Task and attach a Schedule. The Actor remembers which articles of each tag, author, publication and keyword it already delivered (in a Key-Value Store named medium-scraper-only-new, in your own Apify account), so every run delivers, and charges, only the articles it has not given you before; they do not count in maxItems either. Authors and publications stop paging at the first page you already have. A day with nothing new ends green with the status "No new articles since the last run" and costs nothing for results. Point it at a tag with tagSort: latest or at the publications you follow, connect Slack, Google Sheets or a webhook, and new articles land where you read them. Delete the store to start over.

{ "tags": ["artificial-intelligence"], "tagSort": ["latest"], "publications": ["pub.towardsai.net"], "maxItems": 25, "onlyNew": true }

Output fields

Field	What it is
id, title, subtitle, url	The article
publishedAt, updatedAt	ISO 8601 UTC
claps, responses	Engagement as Medium shows it
readingTimeMinutes, wordCount	Length
isMemberOnly	Behind the Medium paywall
tags, imageUrl	Tag slugs and preview image
authorName, authorUsername, authorUrl, authorFollowers, authorBio	The writer's public profile
publicationName, publicationSlug	The publication, when there is one
content, contentMarkdown, contentIsPreview	Body text (with includeContent); contentIsPreview is true for member-only articles
source, rank	Which input produced the row, and its position in search or tag results

Error codes (never charged)

errorCode	Meaning
NO_RESULTS	Nothing for that keyword, tag, author or publication in the date window
NOT_FOUND	No such author or publication, or the article was deleted or made private
INVALID_URL	The line is not a Medium article link or id
INVALID_INPUT	A filter has an unknown value
BLOCKED, NOT_REACHED, UNEXPECTED	Medium did not answer, the run timeout arrived, or something unforeseen broke; only that input is lost

Pricing

Event	Per 1,000
Article delivered	US$ 2.00
Article text (only with includeContent)	US$ 0.80

Duplicates, articles outside the date filter and error rows are free. There is no Apify compute bill on top.

Limitations

• Member-only articles: logged-out readers see a preview, and so does this Actor. The row says so in contentIsPreview.
• Tag feeds stop at 25 per sort order. Medium does not page tag feeds for logged-out readers; combine sort orders, related tags and keyword search for more.
• Author and publication archives are not strictly by date (Medium mixes pinned and featured posts), so publishedAfter skips old ones instead of stopping at the first.
• This Actor reads public pages only and respects the site's terms.

FAQ

Do I need a Medium account or cookies? No. The Actor reads public Medium pages without login or cookies.

What does a run cost? US$ 2.00 per 1,000 articles, plus US$ 0.80 per 1,000 when you turn on includeContent for the full text. Duplicates, articles outside the date filter and error rows are free, and there is no Apify compute bill on top.

Can I get the full text of member-only articles? No. Logged-out readers see a preview, and so does this Actor; those rows have contentIsPreview set to true. Turn on excludeMemberOnly to skip paywalled articles, and they are not charged.

Why does a tag return only about 25 articles? Medium does not page tag feeds for logged-out readers. Pick several orders in tagSort (each one returns up to 25, and an article in more than one feed is charged once), or add related tags and searchQueries.

How do I get only recent articles? Set publishedAfter to a date (YYYY-MM-DD) or a number of days back (30 = the last 30 days). Older articles are skipped and not charged.

Can I schedule it and get only new articles? Yes. Turn on onlyNew, save the input as a Task and attach a Schedule. Articles delivered in earlier runs are skipped, not charged and not counted in maxItems, and a day with nothing new costs nothing for results.

Changelog

• 2026-10-17: FAQ in this page.
• 2026-10-04: onlyNew option for scheduled runs: articles delivered in earlier runs are skipped, not charged and not counted in maxItems.
• 2026-09-30, 0.1: first release. Keyword search, tag feeds in six sort orders, author and publication archives, article links, full text in text and Markdown, date filter, free error rows.