LinkedIn ranks as the second most-cited domain in AI search, behind only Reddit. Semrush's analysis of 89,000 LinkedIn URLs found the platform appearing in 14.3% of ChatGPT Search responses, 13.5% of Google AI Mode responses and 5.3% of Perplexity responses - and for B2B and professional queries specifically, it leads every source across six major AI platforms.

Two studies, run independently and covering nearly 1.4 million AI citations between them, agree on what actually earns a citation and what doesn't. Some of it isn't what conventional LinkedIn advice would predict.

What actually drives LinkedIn citations

Original content, not reshares

Semrush found 95% of cited LinkedIn posts are original content, with reshares accounting for just 5%. OtterlyAI's separate analysis of the same landscape found named individual authors take 91.7% of citations against 8.3% for company pages. Both studies point the same way: models cite the person who wrote the thing, not whoever reposted it.

Length has a sweet spot, and it depends on the format

Semrush found cited articles running 500 to 2,000 words and cited posts running 50 to 299 words. OtterlyAI's numbers land close to the same range: a median of 1,021 words for cited articles and 185 words for cited posts. Word count itself barely matters once you're in the right band - OtterlyAI measured a Pearson correlation of just 0.03 between word count and citations, close enough to zero to ignore. Match the length to the format, not to a target word count.

Educational intent beats promotional

Semrush found 54% to 64% of cited LinkedIn content is knowledge or advice-driven - for Google AI Mode specifically, that rises to almost two-thirds. Content that answers a question outperforms content that pitches something.

Consistency beats a single viral hit

About three-quarters of cited post authors had published five or more times in the four weeks before their post was cited. Engagement told a different story: the median cited post carried 15 to 25 reactions and no more than one comment, and OtterlyAI found Pearson correlations between -0.06 and -0.02 for likes, comments and emojis against citations - functionally no relationship at all. Posting regularly correlates with getting cited. Racking up reactions doesn't.

Individual profiles or company pages - it depends which model you're optimising for

Semrush found a genuine split: Perplexity cites LinkedIn company pages 59% of the time, while ChatGPT Search and Google AI Mode cite individual profiles 59% of the time each. OtterlyAI's wider dataset confirms the individual-authorship lean overall - named individuals took 91.7% of all LinkedIn citations - and found Microsoft Copilot leaning almost entirely on long-form articles, at 90.2% of its citations. Which one to prioritise depends on which platform's audience matters most to you.

Which platforms actually cite LinkedIn

OtterlyAI's analysis of 1.31 million LinkedIn citations found Perplexity responsible for 43.3%, Google AI Overviews 22.2%, ChatGPT 18.7%, Google AI Mode 9.0%, Microsoft Copilot 6.8%, and Gemini close to none at all. Cross-platform citation is rare: 87.4% of cited LinkedIn URLs get picked up by only one platform, though the 2.6% cited by three or more platforms account for 24% of all citations between them. If LinkedIn citations matter to your visibility, Perplexity and Google AI Overviews are doing most of the work.

The myth: hashtags don't control whether you get cited

A common claim is that hashtags shape the URL structure a post gets and improve its odds of citation. OtterlyAI tested it directly: hashtags scored a Pearson correlation of -0.02 against citations - statistically no relationship at all. Their own conclusion: "What gets a post liked does not get it cited."

What this actually means for a B2B posting schedule

The two studies point to a small, specific set of choices rather than a general content strategy:

  • Publish under a real name, not solely from a company page
  • Write full-length articles that answer a question completely, not partial takes that need a comment thread to finish the point
  • Keep a regular publishing cadence rather than chasing one viral post
  • Stop treating hashtags and engagement metrics as levers - the data shows neither moves the number that matters

This sits inside the broader discipline of getting cited in AI search, not just on LinkedIn - for what that work covers day to day, see what an AI search consultant actually does.

A few common follow-up questions

Does follower count matter?

Not much. Semrush found creators with under 500 followers got cited at least as often as accounts with a larger following. A small, specific answer beats a well-followed account with a vague one.

Should companies stop posting from LinkedIn pages altogether?

No. Perplexity still cites company pages 59% of the time. Keep the company account active for Perplexity's audience, and let named individuals publish alongside it for the citation share the other platforms favour.

Do images or video help citations?

No - if anything, the opposite. OtterlyAI found text-only posts averaged more citations than posts with images or video. Media helps a post get noticed by people. It doesn't help a model decide to cite it.