LinkedIn Post Search Scraper — Keywords, No Cookies avatar

LinkedIn Post Search Scraper — Keywords, No Cookies

Pricing

from $0.90 / 1,000 posts

Go to Apify Store
LinkedIn Post Search Scraper — Keywords, No Cookies

LinkedIn Post Search Scraper — Keywords, No Cookies

Search LinkedIn posts by keyword, from $0.90 per 1,000 posts, no cookies, no login. Each post as a row: text, real publish timestamp, author and headline, likes, comments, shares, reactions, media. Filter by newest or top, last 24h/week/month, content type, author. New posts only on a schedule.

Pricing

from $0.90 / 1,000 posts

Rating

0.0

(0)

Developer

Steadyfetch Team

Steadyfetch Team

Maintained by Community

Actor stats

0

Bookmarked

38

Total users

25

Monthly active users

6 days ago

Last modified

Features

Limited permissionsMulti-tenantServer interface

Share

Search LinkedIn posts by keyword — from $0.90 per 1,000 posts, no cookies, no login — and get every post as a row: what was written, when it was really published, the author and their headline, likes, comments, shares, every reaction type, the images, video, document and article attached to it — plus the keyword that found it and its rank. No proxy to configure. Run it on a schedule and it returns only the posts that are new since your last run — posts your account bought in the last 90 days are skipped and not charged again. Click Start with nothing set and it searches a small live sample, and nothing is charged for it.

Using an AI agent? Pin this actor in Apify's MCP server with one link: https://mcp.apify.com?tools=steadyfetch/linkedin-post-search-scraper

  • Actor id: steadyfetch/linkedin-post-search-scraper
  • Input: { "keywords": ["web scraping", "\"AI agents\""] } — the one field you have to set: one search per line, typed the way LinkedIn's own search box reads it. Add "maxPosts": 50 for the per-keyword limit and "maxItems": 1000 for the whole run.
  • Filters LinkedIn applies for you: sortBy (date, the default, or relevance), postedLimit (24h, week, month; sorted by date, a search that comes back empty with it left empty or set to month is searched again over the last week), contentType (videos, images, live_videos, documents, collaborative_articles, jobs), authorKeywords (words on the author's profile, e.g. founder) and authorsCompany (company links or ids).
  • Cap the bill: set maxTotalChargeUsd on the run (a run option, not Actor input), e.g. 1.00 — the run stops when it reaches it.
  • Price: from $0.90 per 1,000 posts on Gold and above, $1.20 per 1,000 on the Apify free plan, plus one $0.006 keyword check for each keyword a run searches. Full table below.
  • How often it changes: a popular keyword gets new posts every few minutes, a niche one a few a week, and like and comment counts keep moving for about 72 hours. A daily run is the shape that fits most watches; an hourly one suits a busy keyword with "Posted in the last" set to 24h.

Unofficial, not affiliated with LinkedIn — not endorsed by or sponsored by LinkedIn or Microsoft Corporation, nor by any author whose public posts appear in a search.

The publish time is the real one

Every row carries postedAt — the moment LinkedIn itself stamped on the post, to the millisecond — and not a reading of a "2 weeks ago" label. You can check it yourself: the timestamp is the same epoch LinkedIn encodes inside the post's own id. postedAgo carries the label too, as a second column, never as the date.

What you get

One row per post.

ColumnWhat it holds
keyword · positionthe search that found the post, and its rank in that search (1 = LinkedIn's first)
postId · postUrl · shareUrl · shareUrnthe post's own ids and its permalink
textwhat was written, in full
links[]the links as written in the post, in order, once each
taggedPeople[] · taggedCompanies[]the people and the companies the post tags, each as { name, url }
postedAt · postedAtTimestamp · postedAgothe real publish moment, and LinkedIn's own label beside it
authorType · authorName · authorHeadline · authorUrl · authorHandle · authorId · authorFollowers · authorAvatarUrl · authorWebsitewho posted it — person or company; their LinkedIn id, one per author, to match their posts across keywords; a company's follower count when LinkedIn states it; the website their profile or page lists, when it lists one
likes · comments · shares · reactionsTotalthe engagement
reactionsLike · reactionsCelebrate · reactionsLove · reactionsInsightful · reactionsSupport · reactionsFunnyeach reaction type as its own column — no list to unpack
countsHidden[]which counts LinkedIn is hiding, so a hidden total never reads as a zero
mediaType · imageCount · imageUrls[] · hasVideo · videoUrl · videoThumbnailUrl · documentUrl · documentTitle · documentPageCount · documentPageImageUrls[]what is attached to the post; for a document, its page count and images of its first pages (documentUrl holds the whole document)
articleTitle · articleUrl · articleDescription · articleSubtitle · articleImageUrl · jobTitle · jobUrl · newsletterTitle · newsletterUrlthe link preview (with its site line and its image), job or newsletter a post shares
isRepost · repostOfPostId · repostOfPostUrl · repostOfPostedAt · repostOfText · repostOfAuthorName · repostOfAuthorUrl · repostOfAuthorHeadline · repostOfImageUrls[] · repostOfVideoUrl · repostOfVideoThumbnailUrl · repostOfDocumentUrl · repostOfDocumentTitle · repostOfDocumentPageCount · repostOfDocumentPageImageUrls[] · repostOfArticleTitle · repostOfArticleUrl · repostOfArticleDescription · repostOfNewsletterTitle · repostOfNewsletterUrl · repostOfLinks[]a repost, with the original underneath it: the original post's id, link and publish time, its own text, its author (name, link and a person's headline), and the images, video, document, article, newsletter and links it carries. On a repost with words of its own, those words stay in text. Empty on a post that is not a repost, and wherever the original carries no such thing
charged · chargedEvents · status · statusReason · retryableon every row, so the invoice reconciles from the dataset

New posts only, on a schedule

  • Every run remembers the posts it delivered to your account, in the key-value store linkedin-post-search-account in your own Apify account. A later run skips them: not delivered again, not charged, and the status line says how many. A post two of your keywords both find is one post, delivered and charged once.
  • Sorted by date (the default), LinkedIn lists a search newest first, so once a page shows posts you already have, the run reads no older page for that keyword. A daily run of a busy keyword reads one or two pages and returns what is new, plus its one keyword check. When a newest-first search with "Posted in the last" left empty, or set to month, comes back empty, the run reads the last week newest first (not when "Posted before" is a week or more back: the last week cannot hold those dates, so it goes straight to the check below). If the last week lists posts, that keyword returns the newest posts from the last week and pays no keyword check, and the run says so on its status line and on one uncharged note row. If the last week is empty too, or was not read, the run looks once at LinkedIn's most-relevant order: no posts there either is a search with no posts on this run, with its keyword check as usual; posts there mean the newest-first list could not be read, so the keyword's row asks you to re-run and nothing is charged for it. When LinkedIn's own count says the search has posts but none came back, the keyword's row asks you to re-run and nothing is charged for it.
  • A keyword that had nothing new for your account is left alone by your runs for up to 24 hours under the same dates, filters and settings (12 hours with "Posted in the last" set to 24h): it gets one uncharged skipped.all_repeat row saying when it was searched and when a run searches it again, and it pays no keyword check. When every keyword of a run is left alone, the run writes one such row for all of them, naming how many and when the first is searched again. Turning on "Include posts you already have" searches a keyword that a run with it off left alone, and changing the dates or the filters searches every keyword now.
  • Turn on "Include posts you already have" to get them again (marked repeat: true, never charged), or set "Posted before" to reach older posts. Posts older than 90 days in your account's memory count as new again. Sorted by relevance, a repeat run cannot stop at what you have, so it reads every page your limit needs.
  • Image, video, document and avatar links are signed by LinkedIn and expire within days. Re-run for fresh links.
  • The memory is a key-value store in your own Apify account, so a run started with a scoped API token in restricted-access mode needs key-value store Read, Write and Create permission, or Actor runs set to Full access; without it the run says so on its status line and cannot skip what you already have.

Price

Two charged events. The store card shows from $0.90/1,000 posts — the cheapest tier.

Apify planPer post1,000 postsKeyword check
Free plan$0.0012$1.20$0.006
Bronze$0.0011$1.10$0.006
Silver$0.001$1.00$0.006
Gold and above$0.0009$0.90$0.006

The keyword check is charged once for each keyword a run searches on LinkedIn and finishes, whatever it finds — new posts, none, or an empty search. Every search costs us the same before a single post is known, so a keyword with one new post pays the same check as a keyword with fifty; more posts never cost you less than fewer. It is never charged for an entry that is not a search (a link, a blank line), a keyword your runs left alone because it had nothing new in the last 24 hours, a search a stop cut before it finished, a search LinkedIn did not answer (an error in place of posts, or a page that came back with fewer than half of its posts even when asked again), a newest-first search that came back empty and then listed posts over the last week, or the bare-Start sample. The row that closes the keyword says when it paid one.

No start fee. No second charge for the media, the article preview, the reaction breakdown or the author — they are columns on the one row you already paid for, and a repost is one post at the same price.

maxTotalChargeUsd is a run option, not an input field, and it is a hard ceiling: the run stops collecting when it reaches it, writes out everything already in hand, and books the rest as rows that name the cap and cost nothing.

On the Apify free plan there is a daily allowance. One account makes at most 25 LinkedIn searches in any 24 hours (each page of up to 50 posts is one search), and all accounts on the Apify free plan share a daily allowance on this actor that resets at midnight UTC. Past either, the run stops cleanly: posts already delivered stay delivered, nothing more is read or charged, and each keyword it did not reach gets an uncharged skipped.free_plan_allowance row naming the reset time in UTC. Paid Apify plans carry no daily allowance.

Honest rows

None of these charges a post, and every one arrives as a row you can read rather than as a missing line. The row's status is the first column; statusReason is the sentence.

statusWhat the row says
no_dataLinkedIn shows no (more) posts for this search with these filters on this run. A finished search pays its keyword check here, and when nothing new came back the row says until when your runs leave the keyword alone.
skipped_date_windowThe search has posts, but none of the ones left fall inside your dates — widen "Posted after" or "Posted before".
skipped.all_repeatThe search reached posts your account already had, so the run read no older page — or it had nothing new for your account in the last 24 hours, so it was not searched again.
skipped.page_capThe read reached the 30 pages this actor reads for one search while LinkedIn was still serving — this actor's own bound.
invalid_keywordLinkedIn would not run the search as written — try fewer or simpler words.
input_errorThe entry is not a search (a link, a blank line, more than 200 characters), or a filter named a value this actor does not recognise — the row names the values that work.
feed_unavailableLinkedIn did not answer for this search on this run, or sent back only part of a page even when asked again. Please re-run — nothing was charged.
limit_reachedThis run reached its limit (a new run fetches the rest), or this actor is paused for now — the row says which.
skipped_max_items · skipped_budget · skipped_deadlineOne of your own limits or the run clock stopped the run first, and the row names which.
skipped.free_plan_allowanceOn the Apify free plan only: the plan's daily allowance on this actor was reached before this keyword. The row names the reset time in UTC.

A keyword that was searched and finished pays its one keyword check on the row that closes it (or on one note row when its whole share was delivered), and that row says so.

What a keyword can be

Plain words (web scraping), an exact phrase in quotes ("AI agents"), or LinkedIn's own AND / OR / NOT (apify AND (actor OR scraper)) — up to 200 characters, one search per line. The same search twice in one run is searched once. A link is refused on its own uncharged row: posts on a person's profile or a company page are LinkedIn profile and company posts, and a group's posts are LinkedIn group posts.

Use it as an API endpoint (Standby)

You can also call this actor like an API: send one keyword and get its latest posts back in the same response, with no run to start and no dataset to read. The address is https://steadyfetch--linkedin-post-search-scraper.apify.actor, and the actor's Endpoints tab lets you try it from the browser.

curl -H "Authorization: Bearer <YOUR_APIFY_TOKEN>" \
"https://steadyfetch--linkedin-post-search-scraper.apify.actor/?keywords=web%20scraping&maxPosts=20"
SettingWhat it does
keywordsOne search per request (also keyword, q or query).
maxPostsHow many of the latest posts to return. 50 if you leave it out; at most 50 per request (25 on the Apify free plan).
sortBy, postedLimit, contentType, authorKeywords, authorsCompany, maxItems, postedAfter, postedBefore, includeRepostsThe same settings as on a run, read the same way.

A POST with the same fields as a JSON body works too. The answer is JSON: rows (the same rows a run writes to its dataset, the run summary last), delivered, charged, and notes for anything that was not returned. Each request is charged exactly as a run with the same input: per post returned, and one keyword check for the search. If the charge for a request cannot be completed, its posts are not sent.

Good to know: the endpoint has no account memory, so it does not skip posts you already have. For a watch that returns only new posts, run the actor on a schedule.

On a Standby request, a post your account already has is charged like any other, and one bought there is not remembered for your runs; the 90-day memory applies to runs.

Other scrapers you might want

Limits and good manners

Only posts LinkedIn shows to a public search are returned. A search is screened for a narrow list of abusive phrases and refused for nothing if it carries one. This actor may stop working if LinkedIn changes how the data is served; if it does, items that fail are never charged.

Support

Something off, or a column you need that is not here? Open an issue on the Issues tab — we usually reply within a couple of hours.

If it earned its keep, a rating helps other buyers find it, and saving the actor keeps it one click away.