LinkedIn Company Posts and Engagement Dataset

Challenge pressure

The client needed every public post from a set of enterprise LinkedIn company pages, each with its date, format and engagement, as data rather than a feed.

The situation

A company page shows its posts as a feed. Comparing how often a brand posts with how much response each post gets needs every post as a row: its date, its format and its likes, comments and shares.

What we built

The scraper covered 14 company pages and wrote one CSV per page, with 14 fields per post. Two pages, Alphabet and Apple, came back empty and Cisco returned only 3 posts, so 12 companies carry data; the gap is stated, not dropped. The brands are public company pages that were scraped, not Van Data Team clients.

What changed

The client got 15,192 posts as rows. The dashboard on this page compares reach with volume: AT&T published the most posts (2,619) but ranks sixth on likes, while Google posted 1,017 times and took 37% of all 16.8 million likes. The league table also changes hands most years, with Google and Amazon trading first place from 2017 to 2022.

Constraints and decisions

Engagement is heavily skewed, so the dashboard reports medians, not means: the mean is 1,105 likes per post while the median company sits at 254, and the top 1% of posts carry 25% of all likes. Video is a minority format, with 2,681 native videos (18%) and 479 embedded YouTube posts. Posts that name individual employees are left out of the rows shown on this site.

Technology stack