LinkedIn Learning Courses Scraper

Search the LinkedIn Learning course catalog by keyword and extract structured data about courses, learning paths, and videos — no login or subscription required.

What it scrapes

• Course title and direct URL
• Instructor name
• Duration (seconds + human-readable, e.g. "2h 15m")
• Difficulty level (Beginner / Intermediate / Advanced)
• Skills taught (array of skill names)
• Description (course overview text)
• Asset type (Course / Learning Path / Video)
• Scraped timestamp

Input

Field	Type	Required	Default	Description
searchQuery	string	Yes	"Python machine learning"	Keyword or skill to search for
maxResults	integer	No	50	Maximum courses to return (1–500)
difficultyLevel	enum	No	All levels	Filter by BEGINNER / INTERMEDIATE / ADVANCED
assetType	enum	No	COURSE	COURSE, LEARNING_PATH, or VIDEO
sortBy	enum	No	RELEVANCE	RELEVANCE or RECENCY
proxyConfiguration	object	No	—	Optional Apify proxy settings

Example input

{
  "searchQuery": "Python machine learning",
  "maxResults": 50,
  "difficultyLevel": "BEGINNER",
  "assetType": "COURSE",
  "sortBy": "RELEVANCE"
}

Output

Each scraped record looks like:

{
  "courseId": "123456",
  "courseUrl": "https://www.linkedin.com/learning/python-for-data-science/",
  "title": "Python for Data Science",
  "instructorName": "Joe Smith",
  "durationSeconds": 7200,
  "durationFormatted": "2h",
  "difficultyLevel": "BEGINNER",
  "skills": ["Python", "Data Science", "Pandas"],
  "description": "Learn Python for data science workflows including data manipulation and visualization.",
  "assetType": "COURSE",
  "scrapedAt": "2024-01-15T12:00:00+00:00"
}

Output fields

Field	Type	Description
courseId	string	Unique course identifier extracted from the LinkedIn URN
courseUrl	string	Direct URL to the course on LinkedIn Learning
title	string	Full course title
instructorName	string	Primary instructor or content creator
durationSeconds	integer	Total course duration in seconds
durationFormatted	string	Human-readable duration (e.g. "2h 15m", "45m")
difficultyLevel	string	BEGINNER, INTERMEDIATE, or ADVANCED
skills	array	List of skill names the course teaches
description	string	Course overview / description
assetType	string	COURSE, LEARNING_PATH, or VIDEO
scrapedAt	string	ISO 8601 timestamp when the record was scraped

Fields that cannot be populated for a given course are omitted (no null values).

Use cases

• Skill gap analysis — find what topics and technologies are covered by popular LinkedIn Learning courses
• Content research — survey what instructors and courses exist in a domain
• Learning recommendation systems — build datasets of course metadata for recommendation engines
• Competitive intelligence — monitor the LinkedIn Learning catalog for a skill category

Limitations

• LinkedIn Learning does not expose enrollment counts or ratings in the public search API; those fields are not included
• Course descriptions and skill tags depend on how the course was catalogued by LinkedIn
• Results may vary slightly by geographic region
• maxResults is capped at 500 per run

FAQs

Does this require a LinkedIn account or login? No. The scraper uses LinkedIn Learning's public search endpoint which does not require authentication.

Does it need a subscription to LinkedIn Learning? No. Course catalog metadata is publicly accessible without a subscription.

Why are some courses missing descriptions or instructor names? Some newer or partner-contributed courses may have incomplete metadata on LinkedIn's end. The scraper only outputs fields that are actually populated.

Can I filter by topic or category? Use the searchQuery field with specific topic keywords (e.g. "project management", "data analysis", "React.js"). Combine with difficultyLevel for more targeted results.

What is a Learning Path? A Learning Path is a curated sequence of courses on LinkedIn Learning. Set assetType to LEARNING_PATH to search specifically for these.

How many results can I get? Up to 500 courses per run. For broader research, run multiple searches with different keywords and deduplicate by courseId.

Explore the rest of the LinkedIn suite

Need a different LinkedIn surface? Pair this actor with any of the others in the LinkedIn Suite — all published under the same publisher and built to share the same cookie format and output conventions.

Actor	What it scrapes
LinkedIn Comments Scraper	All comments + reply threads on a post
LinkedIn Company Employees Scraper	Employee list for any company (by URN)
LinkedIn Company Info Scraper	Company About page (size, HQ, industry, specialties)
LinkedIn Company Posts Scraper	Posts published from a company page
LinkedIn Events Scraper	Events by keyword/URL with full event detail
LinkedIn Hashtag Posts Scraper	Posts ranked under a #hashtag
LinkedIn Jobs Scraper	Job listings via the public jobs-guest API
LinkedIn Jobs Scraper Ultra	Same as jobs-scraper + full detail enrichment
LinkedIn People Search Scraper	People search with every LinkedIn facet (role, company, school, location, etc.)
LinkedIn Post Reactions Scraper	Reactors on a post (name, headline, reaction type)
LinkedIn Post Scraper	Full post (text, media, engagement counts, author)
LinkedIn Post Search Scraper	Posts matching a keyword (with date/author/network filters)
LinkedIn Profile Posts Scraper	All posts/reposts/articles for one profile
LinkedIn Profile Scraper	Public profile fields (name, headline, positions, education, skills)
LinkedIn Profile Scraper Pro	Profile fields + extras (recommendations, organizations, languages)
LinkedIn Profile Scraper Pro Ultra	Pro + premium fields (contact info, followers list when allowed)
LinkedIn Profile Scraper Ultra	Profile + the full upstream dash-120 surface
LinkedIn Profile Search by Name	Search profiles by person name (great for matching CSVs of names)
LinkedIn Schools Alumni Scraper	Alumni list for any LinkedIn school page
LinkedIn Top Content Scraper	Trending / top-engagement posts by topic
LinkedIn User Activity Scraper	Reactions + comments + posts feed for one profile

All actors share the same cookie input format (plain li_at OR full cookies JSON array) and the same omit-empty output convention.