Cracking the LinkedIn Algorithm: What I Learned in 2024

One of the biggest challenges of digital strategy is understanding how search and social algorithms work. Platforms like Google, Facebook, and LinkedIn don’t explicitly share how to create content that drives reach and conversions.

Yet, search and social visibility are essential for running a successful digital strategy. As business leaders, we need to figure out how to leverage these platforms effectively to build and grow our audience.

How Do We Figure It Out?

Through experimentation and data analysis.

We create quality content and see how the algorithm responds. We analyze the data, make changes, and start the process all over again. It’s a cycle of testing and learning—one that can yield significant insights into what works and what doesn’t.

Which brings me to today’s article: an analysis of my LinkedIn content from 2024. The goal? To uncover actionable insights that can help you improve your LinkedIn strategy.

The Analysis

For this article, I reviewed my top 50 LinkedIn posts over a six-month period (Q2 and Q3 of 2024). I analyzed metrics like impressions, engagements (reactions), and comments. I also broke down the data by post type: text, images, videos, and polls.

It’s worth noting that this is just one dataset from one person. While there are statistical flaws in relying on this small sample, the findings still provide a valuable, anecdotal peek behind the curtain of LinkedIn’s algorithm.

So, with that in mind—let’s dive in!

What Drives Reach on LinkedIn?

Let’s start with the big question: what works on LinkedIn?

The answer, as always, is “it’s complicated.”

It depends on the results you’re seeking. Are you aiming for impressions (reach)? Engagements (comments and reactions)? Or conversions (clicks to external landing pages)?

While these goals often overlap, getting reach—or impressions—is one of the most critical outcomes. After all, no one can engage or convert if they don’t see your content.

Here’s what could drive reach on LinkedIn:

1. Comments – The more people who comment on your post, the more LinkedIn’s algorithm seems to favor it. We’re also told the quality of the discussions in the comments matters as well.
2. Engagements – Reactions (likes, loves, etc.) and shares also contribute to your reach. 
3. User Behavior – Actions like clicks to expand text, watch time for videos, and shares via direct messages play a role.
4. Quality and Relevance – LinkedIn’s editorial team has emphasized that relevancy matters. Content aligned with your expertise—based on your skills, experience, and education—gets rewarded.
5. Post Type - The type of content you create, whether its text, image, video, polls, documents, articles, will affect the way your post is reached through the feed. 

The above variables are what we know (through LinkedIn telling us) or believe (through experimentation) impact our ability to reach an audience on LinkedIn successfully. Through the rest of the article, I’ll explain what my data showed about those variables. 

The (Small) Role of Reactions and Comments

One of the most surprising discoveries of my data was that reactions and comments had very little impact on the reach of posts. Like, very little.

Doing some statistical analysis, I looked at the correlation between comments and reactions on posts, and the amount of impressions that post received. 

Here’s what my data showed:

• Comments vs. Impressions - There was an insignificant correlational relationship between the number of comments on a post, and the impressions it received. The R-squared value was 0.017.
• Engagements vs Impression - There was an insignificant correlational relationship between the number of comments on a post and the impressions it received. The R-squared value was 0.061

With an R-squared value of less than .3, my data showed that comments and reactions did not correlate to impressions on my posts. This was honestly very surprising to me, as my assumption was always that engagements on the posts were a major factor in reach. 

But there are a couple of things to keep in mind, even though there was no significant correlation in my dataset:

1. Engagements are still important for conversion. Every comment or like enables you to see who is engaging with your posts and potentially follow-up with a sales activity.
2. Video posts had significantly lower engagement (as we’ll discuss later). This could have impacted the calculation of correlation in my data.

Video: The King of Impressions

If there’s one thing LinkedIn made clear in 2024, it’s this: video is a priority.

The platform introduced a dedicated video tab on its mobile app and multiple video carousels on the desktop feed. These updates dramatically increased video impressions.

For me, video posts dominated. Among my top 50 posts, video averaged 138,000 impressions per post. In comparison, text, image, and poll posts averaged about 2,700 impressions each.

Why the disparity? LinkedIn’s algorithm was actively pushing video content through its new video feed experience. 

The data here is pretty strong - if you want to be seen, you should be experimenting with a video strategy. 

Text and Images: The Engagement Leaders

While videos excelled in impressions, text and image posts outperformed in terms of engagement.

Text posts, in particular, were the most effective at starting conversations and generating comments. These posts tend to invite discussion, making them a great tool for building relationships and encouraging dialogue with your audience.

In fact, my text posts received nearly 100X the amount of comments per impression than my video posts did. This is HUGE. Image posts weren’t far behind at 65X the comment rate of video.

This data tells me that while video posts are the king of impressions, text and image posts could still be the best way to foster engagement through comments and reactions.

It could also be a symptom of user behavior in those watching videos in their feed versus those reading text posts. Does that impact whether or not we comment and engage? Could the quality of video impressions be lower? Maybe so.

Variability: The LinkedIn Wild Card

One major takeaway from my analysis: results will vary.

LinkedIn posts can be highly unpredictable. Some posts receive 500–2,000 impressions, while others soar to 500,000+. 

My top posts were mostly outside the range of “expected” results for my dataset. Using a control chart analysis, it showed that most posts perform “okay,” and a select few are significantly above the upper limits of normal. 

This variability is why consistency is key. The more “at-bats” you have, the greater your chances of hitting a home run. 

You can’t expect most of your posts to perform incredibly well. In fact, a majority of your content will not. Focus on creating valuable content consistently, and over time, you’ll see results.

Five Big Takeaways

In conclusion, here are a few key takeaways that I’ve gathered from my most recent data analysis.

1. Video is a Winner for Impressions - If you want reach, video is the way to go—especially with LinkedIn’s ongoing push for video content. 
2. Variability is Extreme - Don’t get discouraged by low-performing posts. Stay consistent, and the algorithm will reward you over time.
3. Text Posts Drive Engagement - If you’re looking to spark conversations, text posts are your best bet.
4. The Algorithm has MANY variables - What works today may not work tomorrow. Stay flexible and adapt to new trends. User behavior appears to be a more complex variable that we can’t fully see. 
5. Focus on Value and Quality - Align your content with LinkedIn’s goals: creating high-quality, relevant posts that keep users engaged on the platform.