LinkedIn Company Posts Scraper (No Cookies)

Extract public LinkedIn company posts into an Apify dataset using the ScrapeCreators API. This Actor requires no LinkedIn login, session cookies, browser, or proxy settings. The owner supplies the provider key through a private environment secret; users provide company URLs and bounded limits.

What it extracts

Field	Meaning
postId	Stable activity ID for deduplication
postUrl	Public post permalink
companyUrl	Canonical input company page
companySlug	Company URL slug or numeric identifier
text	Public post text, including Unicode
publishedAt	Provider ISO date, or null when unavailable
page	Source page, from 1 to 7
scrapedAt	Collection timestamp in UTC

Use it for competitor content monitoring, marketing research, company announcement tracking, and building a company post dataset for analysis. It does not promise a complete historical archive. Personal profiles, private posts, deleted content, employee lists, reactions, comments, and media enrichment are outside this Actor’s contract.

Input

{
  "companyUrls": ["https://www.linkedin.com/company/shopify/"],
  "maxResults": 20,
  "maxPostsPerCompany": 20,
  "startPage": 1,
  "maxPagesPerCompany": 1
}

companyUrls accepts 1–20 HTTPS LinkedIn company page URLs. Locale subdomains and /posts/ or /about/ tabs are normalized. Duplicate company URLs are fetched once. Invalid URLs and unsupported input properties are rejected before provider work.

maxResults limits all unique posts in the run (1–1000). maxPostsPerCompany limits each company (1–200). The first reached limit wins. Verified Apify free-plan runs are capped at five posts, enforced before requests and before every write. Local runs with an unknown subscription status are not automatically treated as free accounts.

startPage and maxPagesPerCompany select a bounded portion of the public feed. ScrapeCreators documents a maximum of seven pages. The last selected page cannot exceed seven. Start with one page; increase only when additional history is needed. Empty pages and identical page responses stop pagination. Two consecutive companies without new valid posts stop the run with an explicit warning to bound unproductive provider spend. The Actor makes at most three HTTP attempts per selected page, with sequential company processing, short backoff, a 30-second per-attempt timeout, and a three-minute total runtime bound. Company and result limits keep provider work bounded even if no posts are returned.

Output and diagnostics

One dataset row represents one validated, unique post. Deduplication uses postId across all input companies. Invalid post records are skipped with warnings; no fallback posts are invented. publishedAt is supplied by the provider and can be missing or approximate; it is not independently verified against LinkedIn.

The OUTPUT key-value record contains the outcome, delivered count, provider request count, observed provider credits, warnings, effective limit, stop reason, and SDK-observed charged result events. Outcome values distinguish DATA, EMPTY, PARTIAL, REJECTED, and FAILED. Provider authentication, exhausted credits, blocked sources, timeouts, and malformed responses produce explicit diagnostics. A provider failure that prevents all data makes the run fail; valid partial data remains available.

Export the dataset as JSON, CSV, Excel, or through the Apify dataset API. Use the posts dataset view for a readable table. API callers can POST the example input to /v2/acts/{actorId}/runs with an Authorization header. Supply the Apify token through your application’s secret manager; never place it in the input. MCP clients can invoke the Actor with the same input and retrieve the default dataset plus OUTPUT.

Pricing

The intended all-in pay-per-event prices are below. These prices are configured in the private deployment; final verification is recorded in the QA report. PLATINUM and DIAMOND use the GOLD rates. Underlying Apify platform usage is not passed through separately. The automatic start event occurs even if input is rejected, the source returns no posts, or the provider fails. No result event is charged for empty results or invalid records.

Tier	Discount	Start per run	Per delivered post
FREE	0%	$0.02000	$0.01000
BRONZE	2.5%	$0.01950	$0.00975
SILVER	5%	$0.01900	$0.00950
GOLD	20%	$0.01600	$0.00800

At list price, one post costs $0.03, five posts $0.07, and twenty posts $0.22. At GOLD, the same totals are $0.024, $0.056, and $0.176. Minimum total-charge budget is $0.03. The SDK enforces the remaining budget before dataset writes; the Actor checks it before provider work and each write. Synthetic events are never charged manually.

Provider costs are billed to the owner’s ScrapeCreators account independently of Apify. The documented company-posts endpoint and initial live probe consume one credit per page; retries may consume additional credits. A maximum input selects 140 pages and permits 420 HTTP attempts. A charge cap limits event charges, not the provider’s invoice. The runtime and request bounds apply separately.

Pricing assumes a conservative $0.002 per provider credit, based on the provider’s published $10/5,000 entry pack, subject to the owner’s actual credit plan. Final cloud evidence records measured platform usage and provider credits; prices must be revisited if those assumptions or provider prices change. Apify’s documented creator share is 80% of event revenue, before platform and external provider costs.

Owner setup and reliability

Install with npm ci, run npm test and npm run check, then configure SCRAPECREATORS_API_KEY. For local runs, load it through a secret manager or environment. For deployment, .actor/actor.json references the current user’s configured Apify CLI secret of the same name. No key is stored in source or accepted as a user input field.

The implementation calls GET https://api.scrapecreators.com/v1/linkedin/company/posts with company URL and page number, using x-api-key authentication. Missing or invalid operator credentials are configuration failures, not empty feeds. HTTP 401, 402, 403 and 404 are not retried; transient HTTP errors and network failures receive at most two retries. A circuit breaker stops systemic failures after two company request failures. Run results reflect public availability at collection time.

Development and preset Tasks remain private. Publication is an owner action after checking final cloud evidence, pricing, permissions, branding, and claims. No third-party logo is included. There is no Standby server and no requirement to store LinkedIn sessions.

Responsible use

Collect only public information you are authorized to process, and follow LinkedIn, ScrapeCreators, and Apify terms and applicable data-protection requirements. This Actor cannot access content hidden behind login or other access controls. Retain only the data needed for your research and downstream workflow.

Sources

• ScrapeCreators API specification
• Apify pay-per-event documentation