🚀 LinkedIn Post Reactions Scraper Pro

Extract detailed reaction data from any LinkedIn post — including reactor names, headlines, profile URLs, and photos. Supports batch processing of up to 1,000 posts in a single run.

No LinkedIn account or cookies required.

💡 For other LinkedIn tools by Dende Labs, check: https://apify.com/dendelabs

---

✨ Key Features

Feature	Description
🔄 Auto-pagination	One run extracts every reaction LinkedIn exposes. No manual page management. See the LinkedIn Reactions Limit section for details on posts with very high engagement.
📦 Batch processing	Scrape up to 1,000 posts in a single run.
🔍 Headline keyword filters	Only include reactors matching specific keywords (e.g., "CEO", "VP").
🏢 Company detection	Each reactor is flagged as person or company page.
📊 Incremental tracking	Compare with previous runs — each reactor is marked as new or existing.
🚫 Exclude companies	Filter out company pages, keep only people.
🔗 All URL formats	Paste any LinkedIn post link — even with tracking parameters.
🔐 No cookies needed	No risk of account restrictions or bans.

---

📥 Input

Basic Usage

Just paste one or more LinkedIn post URLs:

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_no-one-becomes-a-clinician-to-do-paperwork-activity-7302346926123798528-jitu"
  ]
}

All Input Parameters

Parameter	Type	Required	Default	Description
postUrls	string[]	✅ Yes	—	LinkedIn post URLs or activity IDs (up to 1,000)
reactionType	string	No	ALL	Filter: ALL, LIKE, PRAISE, EMPATHY, APPRECIATION, INTEREST, ENTERTAINMENT
maxReactions	integer	No	0	Max reactions per post. 0 = unlimited
headlineKeywords	string[]	No	[]	Only include reactors whose headline contains these keywords
excludeCompanies	boolean	No	false	Exclude company/org pages from results
previousDatasetId	string	No	—	Dataset ID from previous run for incremental tracking
metadataOnly	boolean	No	false	Only fetch reaction counts, not individual reactions (see Metadata Only Mode)

Supported URL Formats

All of these work — just paste whatever you have:

✅ https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu
✅ https://www.linkedin.com/feed/update/urn:li:activity:7302346926123798528/
✅ https://www.linkedin.com/feed/update/urn:li:share:7302346926123798528/
✅ https://www.linkedin.com/posts/user_slug-activity-123/?utm_source=share&utm_medium=...
✅ 7302346926123798528

---

📤 Output

The dataset contains two record types:

1. Reaction records — one per reactor (the main output, described below).
2. Summary records — one per post (type: "summary"), pushed at the end of each post's scrape with totals and a truncation flag. See Summary Record and the LinkedIn Reactions Limit section.

If your integration iterates the dataset, filter by the type field (or check for the absence of reactor):

const reactions = dataset.filter(r => r.type !== 'summary' && r.type !== 'error');
const summaries = dataset.filter(r => r.type === 'summary');

Single Reaction

Each reaction in the dataset looks like this:

{
  "reaction_type": "LIKE",
  "is_company": false,
  "reactor": {
    "id": "ACoAAB8v74YB5oqaAg-4F2VRFh9tEt0zXfRsjpE",
    "name": "Ashish Pandey",
    "headline": "Head of Ops @ Wiz Labs",
    "linkedinUrl": "https://www.linkedin.com/in/ACoAAB8v74YB...",
    "profile_pic": "https://media.licdn.com/dms/image/v2/..."
  },
  "_metadata": {
    "post_url": "https://www.linkedin.com/posts/satyanadella_...",
    "activity_id": "urn:li:activity:7302346926123798528",
    "extracted_at": "2026-03-30T12:00:00.000Z"
  }
}

Company Reactor Example

When a company page reacts to a post, is_company is true:

{
  "reaction_type": "LIKE",
  "is_company": true,
  "reactor": {
    "id": "106345960",
    "name": "Dende Labs",
    "headline": "2 followers",
    "linkedinUrl": "https://www.linkedin.com/company/dendelabs/",
    "profile_pic": null
  },
  "_metadata": {
    "post_url": "https://www.linkedin.com/posts/...",
    "activity_id": "urn:li:activity:7399183215997108224",
    "extracted_at": "2026-03-30T12:00:00.000Z"
  }
}

Summary Record (one per post)

At the end of every post's scrape, the Actor pushes one extra record with type: "summary". It reports how many reactions the post actually has, how many were extracted, and whether the result was truncated by LinkedIn's ~1,200-reactor cap.

{
  "type": "summary",
  "post_url": "https://www.linkedin.com/posts/...",
  "total_reactions_on_post": 3605,
  "reactions_extracted": 1245,
  "reactions_matching_filters": 1245,
  "truncated": true,
  "truncation_reason": "linkedin_reactions_list_limit",
  "truncation_note": "LinkedIn limits the public reactions list to approximately 1,200-1,250 reactors per post...",
  "extracted_at": "2026-04-18T00:05:28.000Z"
}

For posts where all reactors were extracted, truncated: false and truncation_reason/truncation_note are null.

Output Fields Reference

Reaction record:

Field	Type	Description
reaction_type	string	LIKE, PRAISE, EMPATHY, APPRECIATION, INTEREST, or ENTERTAINMENT
is_company	boolean	true if the reactor is a company page, false if a person
is_new	boolean	Only present when using incremental tracking. true = new reactor, false = seen before
reactor.id	string	LinkedIn profile or company URN
reactor.name	string	Full name of the person or company
reactor.headline	string	LinkedIn headline / tagline
reactor.linkedinUrl	string	Direct link to the reactor's LinkedIn profile
reactor.profile_pic	string	URL to profile photo (800x800) or null
_metadata.post_url	string	The original post URL you provided as input
_metadata.activity_id	string	LinkedIn activity URN of the post
_metadata.extracted_at	string	ISO timestamp of when the data was extracted

Summary record (type: "summary"):

Field	Type	Description
type	string	Always "summary" — use to distinguish from reaction records
post_url	string	The post this summary applies to
total_reactions_on_post	integer	Total reactions the post has on LinkedIn
reactions_extracted	integer	How many reactors the Actor actually extracted
reactions_matching_filters	integer	Of those extracted, how many match your headline/reaction-type filters
truncated	boolean	true if LinkedIn's ~1,200-reactor cap was hit; false if all reactors were extracted
truncation_reason	string | null	"linkedin_reactions_list_limit" when truncated, null otherwise
truncation_note	string | null	Human-readable explanation when truncated, null otherwise
extracted_at	string	ISO timestamp of when the post finished scraping

---

📌 Examples

Example 1: Get all reactions from a single post

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_no-one-becomes-a-clinician-to-do-paperwork-activity-7302346926123798528-jitu"
  ]
}

Example 2: Batch — scrape reactions from multiple posts

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu",
    "https://www.linkedin.com/posts/billgates_activity-7300000000000000000-abcd",
    "https://www.linkedin.com/feed/update/urn:li:activity:7298765432109876543/"
  ]
}

Example 3: Only get reactions from decision-makers

Use headlineKeywords to filter reactors by their LinkedIn headline:

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"
  ],
  "headlineKeywords": ["CEO", "CTO", "VP", "Founder", "Director", "Head of"]
}

Only reactors whose headline contains at least one of these keywords will be included.

Example 4: Only people, no company pages

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"
  ],
  "excludeCompanies": true
}

Example 5: Filter by reaction type

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"
  ],
  "reactionType": "PRAISE"
}

Available types: ALL, LIKE, PRAISE (Celebrate), EMPATHY (Love), APPRECIATION (Insightful), INTEREST (Curious), ENTERTAINMENT (Funny)

Example 6: Limit results per post

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"
  ],
  "maxReactions": 500
}

Example 7: Incremental tracking — detect new reactors

First run — scrape all reactions:

{
  "postUrls": ["https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"]
}

Second run — pass the previous dataset ID to tag each reactor as new or existing:

{
  "postUrls": ["https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"],
  "previousDatasetId": "abc123xyz"
}

Each reactor in the output will have an is_new field:

• true — this reactor was NOT in the previous dataset (new engagement)
• false — this reactor was already in the previous dataset

💡 You can find the dataset ID in the Apify Console under the Storage tab of your previous run.

Example 8: Metadata only — check reaction count before scraping

Use metadataOnly to quickly check how many reactions a post has without extracting them all. This makes just one API call per post and returns only the totals.

{
  "postUrls": [
    "https://www.linkedin.com/posts/satyanadella_activity-7302346926123798528-jitu"
  ],
  "metadataOnly": true
}

Output:

{
  "post_url": "https://www.linkedin.com/posts/satyanadella_...",
  "total_reactions": 847,
  "total_pages": 9,
  "extracted_at": "2026-03-31T12:00:00.000Z"
}

When to use this:

• Avoid unnecessary scrapes — Check if a post has new reactions since your last run before paying for a full extraction. For example, if you scraped 847 reactions yesterday and the metadata still shows 847 today, you can skip the full scrape.
• Cost estimation — Know exactly how many reactions (and how much it will cost) before committing to a full run.
• Monitoring dashboards — Track reaction growth over time without extracting every reactor each time.

💡 Metadata-only mode costs just $0.004 per post (one result in the dataset). A full scrape of 1,000 reactions would cost $4.00. Use metadata to decide if the full scrape is worth it.

---

🤖 Integration with AI Agents

This actor is optimized for use with AI agents and LLMs via the Apify MCP server.

Output Schema

The actor produces a flat dataset. Most items are reaction records; one summary record (type: "summary") is pushed at the end of each post. Filter item.type !== 'summary' to get only reactions. Key fields on reaction records:

• reactor.name — Who reacted
• reactor.headline — Their professional context
• reaction_type — How they reacted
• is_company — Whether it's a company or person

Example Agent Prompt

"Scrape all reactions from this LinkedIn post and find decision-makers (CEOs, VPs, Directors) who engaged with it. Return their names and LinkedIn URLs."

The agent can use headlineKeywords: ["CEO", "VP", "Director"] to filter directly at scrape time, reducing post-processing.

---

🎯 Use Cases

• 🔎 Sales prospecting — Find decision-makers engaging with competitor or industry content
• 📊 Market research — Analyze who engages with thought leaders in your space
• 🏆 Competitor analysis — Track engagement patterns on competitor posts
• 📋 Lead generation — Build targeted lists from post engagement data
• 📈 Content strategy — Understand what types of professionals engage with specific topics
• 🔄 Engagement monitoring — Track reaction growth over time with incremental mode

---

ℹ️ Profile URL Format

LinkedIn returns profile URLs in profile ID format for reactions, for example:

https://www.linkedin.com/in/ACoAAB8v74YB5oqaAg-4F2VRFh9tEt0zXfRsjpE

Instead of the human-friendly slug format:

https://www.linkedin.com/in/john-doe

This is a LinkedIn limitation — the reactions popup only exposes profile IDs, not public slugs. Both formats are valid and redirect to the same profile. This is the standard behavior across all LinkedIn reaction scrapers.

---

⚠️ LinkedIn Reactions Limit (~1,200 per post)

LinkedIn caps its public reactions list at approximately 1,200-1,250 reactors per post, no matter how many reactions the post actually has. If you scrape a post with 3,000 reactions, only the first ~1,200 reactors are exposed by LinkedIn's public list — the rest are invisible to any scraper that does not use a logged-in LinkedIn account cookie (li_at).

This is a LinkedIn platform limitation, not a limitation of this Actor. Every cookieless reactions scraper is subject to the same cap.

How this Actor handles it:

• Every post's output ends with a summary record (type: "summary") that includes total_reactions_on_post, reactions_extracted, and a truncated flag — see the Summary Record section for the full schema.
• The Actor also logs a WARNING in the run log when truncation is detected, so it's visible without parsing the dataset.
• Use Metadata Only mode to check the true total reaction count on a post before running a full scrape, with no per-reaction cost.

What this means for you:

• For posts with ≤ 1,200 reactions (the vast majority of posts), you get the full list.
• For high-engagement posts (viral content, posts from top influencers), the Actor extracts the first ~1,200 reactors and reports the truncation transparently — no silent data loss.

---

⚠️ Disclaimer

This Actor is an independent tool and is not affiliated with, endorsed by, or sponsored by LinkedIn Corporation. LinkedIn® is a registered trademark of LinkedIn Corporation. All trademarks are property of their respective owners.