LinkedIn Pulse Search Scraper — Keyword filter across curated authors

Keyword-search LinkedIn Pulse long-form articles across a curated list of authors you specify. For each author, the actor scans their Pulse archive and returns articles matching your keyword(s) in title or description. Clean structured JSON output. No LinkedIn login, no cookies, no browser session, no account ban risk.

Built for competitive content intelligence, thought-leadership monitoring, AI training-data curation by topic, and content-trend tracking across a known audience.

---

Why curated-author search? — being honest about scope

Most LinkedIn search scrapers target all of LinkedIn. We tried to build that for Pulse — but LinkedIn auth-gates Pulse-specific search and discovery surfaces (/pulse/discover, /pulse/topic/*), and the vendor /search/posts endpoint returns only short-form posts (zero Pulse hits across every keyword we tested).

So we built something honest instead: curated-author search. You supply the authors you care about (CEOs, analysts, thought leaders); we scan their archives + filter by keyword. Narrower than "all of LinkedIn Pulse" but:

• ✅ Works today, fully cookieless
• ✅ Cheap: $1/1k matching hits
• ✅ Predictable cost (you control how many authors × how many articles to scan)
• ✅ Ideal for monitoring known thought leaders, not random discovery

---

What you get — 1 record per matching article

For each author in profileUrls, the actor:

1. Resolves the profile URN
2. Paginates their Pulse archive (/articles/all)
3. Filters each article by keyword match against title + description
4. Emits a pulseSearchHit record for every match (up to maxHitsPerAuthor)

pulseSearchHit record (13 fields)

• Matching info: matchedTerms (which keywords hit), positionInArchive (0 = most recent)
• Article: articleUrn, navigationUrl, title, description, duration, durationMinutes, coverImageUrl
• Author: authorUsername, authorProfileUrl
• Tracking: sourceInput, scrapedAt

Plus diagnostic (1 per author with no matches) and summary (1 per run).

Match modes

Mode	Behavior	Example
any (default)	Article matches if ANY keyword hits	Keywords "AI, agents" → matches articles containing AI OR agents
all	Article matches only if ALL keywords hit	Keywords "AI, agents" → matches only articles containing both AI AND agents

Use cases (top 6)

1. Track competitor CEO thought-leadership by topic

Feed 10-20 competitor CEO URLs + your category keyword ("AI agents", "climate tech", etc). Daily cron run surfaces every new article in the topic — alerts you to thematic content shifts in real-time.

2. AI training data — curate domain-specific content

Feed 100 experts in your domain + topic keywords. Filter to articles where they wrote on your topic. Use navigationUrl to chain to the Pulse Article Scraper for full HTML content.

3. Content marketing intelligence — what's resonating in your category

Compare keyword hit-rates across thought leaders. An author with 50% AI-articles signals investment depth. Useful for partnership scoring.

4. Recruiting signal — domain depth via published content

Candidates who consistently publish on a topic (high hit-rate matching the topic) demonstrate domain depth. Useful for senior+ roles.

5. Investor research — founder thesis-tracking

Feed portfolio CEO URLs + thesis keywords. See how their public writing evolves over time on the topics that matter.

6. PR + analyst monitoring

Track named analysts/journalists' Pulse output for mentions of your company, product, or category. Surfaces commentary you'd miss in mainstream press.

How it works

1. Set profileUrls — LinkedIn profile URLs (https://www.linkedin.com/in/<username>/) or bare usernames whose archives to search.
2. Set keywords — comma-separated terms to match against article title + description.
3. Choose matchMode (any or all).
4. Set maxArticlesToScanPerAuthor (1-500, default 50) and maxHitsPerAuthor (1-500, default 50).

For each author: resolve URN → paginate /articles/all (20 per page) → filter each article → emit matching hits.

No /articles/article/info calls — just /urn + /articles/all. No vendor cost per matched article beyond the page-pagination.

No login required. No cookies. Cookieless via managed vendor partnership — no account ban risk.

Example input

{
  "profileUrls": [
    "https://www.linkedin.com/in/satyanadella/",
    "williamhgates",
    "reidhoffman"
  ],
  "keywords": "AI, agents, future",
  "matchMode": "any",
  "maxArticlesToScanPerAuthor": 50,
  "maxHitsPerAuthor": 20
}

For 3 authors × 50 articles scanned × matching ~20% → roughly 30 hits → cost: $0.03 total at $1/1k.

Pricing — pay only for matching hits

$1.00 per 1,000 matching hits. Pay-per-result.

• ✅ No subscription
• ✅ No charge for non-matching articles (we scan them for free; only matching hits are billed)
• ✅ No charge for authors with zero matches (clean diagnostic record)
• ✅ Apify's free tier ($5/month platform credit) covers ~5,000 hits to start
• ✅ Predictable cost: capped by maxHitsPerAuthor × number_of_authors

Cost comparison:

Tool	1,000 hits	Pulse-aware?
LinkedIn Search Scraper (a20) POSTS entity	$1.50/1k	❌ (returns regular posts only, no Pulse content)
harvestapi / supreme_coder / others	—	❌ (no Pulse search)
This actor	$1.00/1k	✅ explicit Pulse

Combine with other LinkedIn actors

Build a multi-stage pipeline:

1. This actor (p6) → filter by keyword across N authors, surface matching articles
2. LinkedIn Pulse Article Scraper (p1) → pipe navigationUrl for full HTML + engagement on selected hits
3. LinkedIn Pulse Comments Scraper (p2) → comments on high-engagement matching articles
4. LinkedIn Pulse Reactions Scraper (p3) → reactor lists for content-intent signals
5. LinkedIn Profile Scraper (a17) → enrich the article authors with full firmographic data
6. LinkedIn Pulse Author Archive (p4) → if you need to expand the curated-author list before searching

FAQ

Q: Can I search ALL of LinkedIn Pulse without specifying authors? No — that requires LinkedIn's auth-gated search APIs which no LinkedIn-data vendor exposes today. Build a curated author list from LinkedIn Search Scraper (a20) with entityType=PEOPLE and a topic-related title filter, then feed those usernames here.

Q: Why does the match check only title + description, not full content? Article body HTML is only available via the more expensive /articles/article/info call ($2/1k via Pulse Article Scraper). We do title + description matching cheap, then customer chains to p1 for content matching on selected hits.

Q: How are keywords combined when matchMode=any vs all? ANY: matches if at least one keyword appears. ALL: matches only if every keyword appears. Use ALL for narrow targeted searches, ANY for broader topic monitoring.

Q: Are matches case-sensitive? No — all matching is case-insensitive. AI will match ai, AI, Ai, etc.

Q: What if an author has 0 articles? The actor emits a diagnostic record for that author and moves on. Other authors in the batch proceed normally.

Q: Can I see WHICH article positions matched? Yes. Each pulseSearchHit has positionInArchive (0 = most recent). Useful for trend tracking.

Q: Is this safe to use vs my LinkedIn account? Yes. No login, no cookies. The actor never authenticates as you.

Privacy & compliance

This actor scrapes publicly accessible LinkedIn Pulse article listings — the same titles + descriptions any logged-out visitor can see on a profile's articles tab.

LinkedIn content may be subject to LinkedIn's Terms of Use and the article author's copyright. Customers are responsible for compliance with applicable laws (copyright, GDPR, CCPA).

Not affiliated with, endorsed by, or sponsored by LinkedIn Corporation or Microsoft Corporation.