LinkedIn Top Content & Top Voices Directory Scraper

Scrape LinkedIn's public Top Content directory and its quarterly Top Voices class articles. The directory is LinkedIn's own editorial collection of topic pages — each page ranks 9–10 posts from recognized professionals — and the Top Voices articles list the honorees of each class. No login, cookie, API key, or proxy setup is required.

What this actor does

• Three modes: topContent for the curated topic directory, topVoices for quarterly class lists, and postUrls to enrich specific public posts.
• Depth 2–4: scrape only the selected topic pages, add their subtopics, or add the leaf pages beneath those subtopics.
• Exact dates and full text: every post carries its real ISO publish date, full text, first-line headline, author, follower count, avatar, media, topic path, and editorial rank.
• Top comments: the public post page embeds up to 10 comments with author, date, text, and reactions.
• Filters and sorting: date windows, exact date bounds, media type, minimum reactions/comments, Influencer-badged authors, and five sort orders.
• Honest output: unavailable fields are omitted, never filled with zero or a placeholder. Duplicate posts across topics are delivered once.

Output per post (mode = topContent or postUrls)

• recordType — post
• postId — LinkedIn activity ID
• postUrl — canonical public post URL
• headline — the post's first line
• content — full post text
• postedAt — exact ISO 8601 publish timestamp
• authorName, authorProfileUrl, authorAvatar
• authorHeadline — author subtitle when the page shows one
• authorFollowers — follower count shown by LinkedIn
• authorIsInfluencer — whether the public author card carries the Influencer marker
• authorBadge — Influencer when that marker is present
• reactionsCount, commentsCount — real served counts
• comments[] — up to maxComments embedded comments, each with authorName, text, postedAt, and reactionsCount
• mediaType — text, image, video, or article
• mediaUrls[], mediaItems[] — image/video URLs, plus thumbnail, duration, and caption URL for videos
• articleTitle, articleUrl — shared article or newsletter link
• linkPreview — {url, title} for shared article links
• hashtags[], mentions[], externalUrls[] — parsed from the post text
• topicPath, topicTitle, topicCategory, topicUrl — the directory location that surfaced the post
• topicRank — 1-based position on that topic page
• engagementScore — reactions + 3 × comments
• scrapedAt — ISO 8601 timestamp of the run

Output per Top Voices honoree (mode = topVoices)

• recordType — topVoice
• voiceName, voiceProfileUrl
• voiceHeadline — when the honoree card shows one
• voiceFollowers — follower count on the card
• voiceBadge — Influencer when the card carries the marker
• voiceClass — class label such as September 2026 US
• voiceRegion — US or Global when the class title identifies it
• sourceUrl — the class article URL
• scrapedAt

Input

Field	Type	Default	Description
mode	string	auto	auto, topContent, topVoices, or postUrls. Auto picks post URLs first, then Top Voices class URLs, then the directory.
topics	array	–	Topic or subtopic paths selected from the picker, or custom /top-content/... paths and URLs.
maxDepth	string	2	2 = selected topic pages only; 3 = add subtopics; 4 = add leaf pages beneath subtopics.
startUrls	array	–	Exact Top Content pages to scrape instead of expanding topics.
postUrls	array	–	Public post, activity, share, or ugcPost URLs to enrich.
topVoiceClassUrls	array	–	LinkedIn News Top Voices class articles; leave empty for the latest US class.
maxVoices	integer	100	Maximum honorees after deduplication.
maxPosts	integer	50	Maximum posts delivered after deduplication, filters, and sorting.
sortBy	string	linkedin_rank	linkedin_rank, newest, most_reactions, most_comments, or most_engagement.
dateRange	string	–	24h, 7d, 30d, 90d, or 1y; leave empty for any time.
postedAfter	string	–	Inclusive lower date bound (YYYY-MM-DD).
postedBefore	string	–	Inclusive upper date bound (YYYY-MM-DD).
minReactions	integer	0	Minimum reactions.
minComments	integer	0	Minimum comments.
mediaTypeFilter	string	–	text, image, video, or article.
onlyInfluencers	boolean	false	Keep only authors with the public Influencer marker.
includeComments	boolean	true	Attach the embedded comments.
maxComments	integer	10	Comments to keep per post (1–10; public pages embed about 10).
includeVideoTranscript	boolean	false	Add a transcript when the public video publishes one.
proxyConfiguration	object	–	Optional override. The actor works without a proxy.

Example: a quick topic sample

{
  "mode": "topContent",
  "topics": ["artificial-intelligence"],
  "maxDepth": 2,
  "maxPosts": 10,
  "includeComments": true
}

Example: deeper crawl with filters

{
  "mode": "topContent",
  "topics": ["leadership", "future-of-work"],
  "maxDepth": 3,
  "maxPosts": 200,
  "sortBy": "most_engagement",
  "dateRange": "1y",
  "minReactions": 50,
  "onlyInfluencers": true,
  "includeComments": false
}

Example: exact topic pages

{
  "mode": "topContent",
  "startUrls": [
    "https://www.linkedin.com/top-content/artificial-intelligence/advancing-ai-development/",
    "https://www.linkedin.com/top-content/career/resume-tips/"
  ],
  "maxPosts": 20
}

Example: enrich specific posts

{
  "mode": "postUrls",
  "postUrls": [
    "https://www.linkedin.com/posts/williamhgates_in-many-parts-of-the-world-the-standard-activity-7462665106229727232-ud0G",
    "https://www.linkedin.com/feed/update/urn:li:activity:7504986609365352448/"
  ],
  "maxComments": 5
}

Example: a Top Voices class

{
  "mode": "topVoices",
  "topVoiceClassUrls": ["https://www.linkedin.com/pulse/meet-september-2026-us-top-voices-experts-leaders-creators-d0tjc"],
  "maxVoices": 100
}

Use cases

• Content research: see which posts LinkedIn's editors rank first in a topic, with exact dates and full text.
• Influencer discovery: collect authors, follower counts, and public Influencer badges from a niche.
• PR and media prospecting: build a list of Top Voices honorees with profile URLs and headlines.
• Benchmarking: compare reactions, comments, and media formats across topics and time windows.
• Corpus building: export complete posts with their top comments for analysis or RAG pipelines.

Limitations and data freshness

• The directory is curated, not a live trending feed. In the October 2026 research sample the newest post was dated 2026-09-18. Posts are often weeks or months old, and a 24-hour or 7-day filter is frequently empty. Every record carries its exact postedAt, and the run status reports the newest date found.
• Each topic page ranks 9–10 posts. There is no public pagination; use maxDepth and more topics, subtopics, or startUrls to reach more posts.
• Comments are capped at the roughly 10 embedded on the public post page. Commenter profile URLs are not published in that data.
• Public fields vary by post. Author headlines appear on about two thirds of sampled posts, and the Influencer marker on about seven in ten. Avatar and follower count are present on every sampled post.
• Some fields are not published on the public surface: repost counts, reaction-type breakdowns, view counts, and company-tagged mentions are omitted rather than fabricated.
• Top Voices class layouts differ. The US class cards expose name, headline, followers, badge, and profile URL; some global class articles expose the honorees named in the article body with name and profile URL only. Comment-section and recommended profiles are not counted as honorees.
• A depth-4 crawl is large. The directory has roughly 16,000 leaf pages; set maxPosts and expect a long run when expanding a topic all the way down.
• Public pages and LinkedIn's terms. This actor reads only public pages and never logs in. LinkedIn's robots policy restricts automated access; you are responsible for how you use the data. This actor is not affiliated with or endorsed by LinkedIn.

FAQ

Do I need a LinkedIn account, cookie, or API key? No. Every mode reads public pages. Old tasks that still send a cookie or language value keep running; those values are ignored and mentioned in the run status.

Is this a live "trending now" feed? No. It is LinkedIn's editorial Top Content directory, which is updated on LinkedIn's schedule. Check postedAt rather than assuming today's date.

Why did I get fewer posts than maxPosts? A topic page holds only 9–10 posts, duplicates across topics are removed, and your filters may remove more. Increase maxDepth, add topics or startUrls, or loosen the filters.

Why is my date filter empty? The curated directory often has no posts inside 24 hours, 7 days, or 30 days. Try 90d or 1y, or leave the date filter empty.

Are the comments complete? No. Public post pages embed about 10 comments; the actor delivers those, up to maxComments, and does not invent the rest.

What does authorIsInfluencer mean? It reflects the public Influencer marker on the author card. It is not a claim that the author belongs to a current Top Voices class — use the topVoices mode for class lists.

Why are reposts, reaction breakdowns, or views missing? Those values are not published on the public pages. They are omitted instead of being returned as zero.

Can I scrape a subtopic or leaf page directly? Yes. Put any /top-content/... path in topics or a full URL in startUrls.

What is the difference between topContent and topVoices? topContent returns ranked directory posts. topVoices returns honoree records from LinkedIn News class articles, using the voice* fields.