Reaching 1.18M impressions on LinkedIn over the last year - A Commentary and Review
From December 3, 2022 to December 3, 2023, I reached 1.18M people through LinkedIn and grew my following by 4827 people. I'm currently at 5299 followers so this seems like a bit of a sus number to me - I guess I only had 472 connections/followers at this time last year.
TL;DR
Post consistently, and don't expect virality
Increase average views by collaborating
Memes are the best statistical spread
Images > Text > Video (mostly)
Average engagement rate of 1%
Statistical Review of Impressions
1.18M impressions over 365 days is about 3238 views a day. Some posts reached way more people than others. My most wide reaching post reached 235,100 people, then 145,262, then 92,346. This means that my top three posts accounted for 472,708 of my 1.18M impressions. That's almost exactly 40% (39.99% lol)!
As you can see from the image below that these posts were made 6 months, 1 year, and 4 months ago, respectively. The second most popular post reached ~61% as many people as the most popular, and the third reached ~64% as many people as the second.
Double clicking into "show more", we can see the next set of five posts as well. The next five posts sum to a total of 104,934 views. That's barely more than the third most popular post, and much less than the second. We can also see a steeper drop off between the third and fourth posts than before. The fourth most popular one is less than 50% as popular as the third.
Statistical Review of Engagements
12,995 engagements for 1.18M impressions is an engagement rate of just over 1%. If you're familiar with marketing and sales funnels, this is about on par with what you'd expect. One interesting point is that you'll see my engagements are up nearly 200% from the year before, but my impressions were only up 100% from the year before. This means that last year, I posted much more engaging content than from 2021 to 2022.
Looking at the top three most engaging posts we see something interesting. They're all the same, but the order of the first two are swapped. And it's not even close. 2008/145,262 is ~1.4% engagement. Meanwhile, 1472/235,100 is only ~0.6%. The Drake meme was much more engaging, more than 2x as much as the one about programming languages.
More on what makes content "engaging" later.
Looking at the next 5 most engaging posts, we see that these are slightly different than the 5 most impressionable posts. Four out of five are the same, but they are in different orders. Post 4 comes in at ~0.6%. Number 5 comes in at ~1.3%, number 6 is ~1.2%. The open source post is at ~0.05%.
Finally, the number 8 most engaging post was one that didn't even make it to the top most impressionable posts. It had 6649 impressions, 106/6649 comes out to a whopping ~1.6% engagement rate.
The difference between the most and second most engaging was not as high as the rate for impressions. #2 was 73.3% of #1. Maybe because the second most engaging post had more impressions. The second to third drop off was much steeper with third place coming in at only 26% of second place. From there it gets more chill.
Which Post "Types" were most popular?
Images by far. Then text. Then video. I had one video make it into the top 8 for impressions, and none for engagements.
The type of images that were the most popular? Memes.
The type of text posts that were the most popular? Callouts and collaborations.
Why was the Drake meme so engaging?
I think that was the most engaging post, while also having a wide reach, because it was a) relatable, b) causes some emotion, and c) familiar. The meme was about spending more time to automate a task that is easily done manually.
As software devs, we know we all do that (relatable). However, some people were upset about it and defended their desire to want to automate things (cause emotion). Finally, it's the Drake meme, most of Gen Z and Millenials have seen that meme somewhere else.
A deeper look into my top posts by engagements and impressions below.
Writing & Marketing @ Cyfrin: World-class web3 education, tools, and security audits | "Metaverse AI" trilogy author
1yAbsolutely brilliant post Yujian Tang. Helpful reality check that ~1% engagement is normal. What are your thoughts on? - number of posts per day/week - commenting and engaging with other creators' posts - human vs AI-generated content in your posts Would be grateful for your thoughts.
Cut Data Stack Cost | dbt + Snowflake + Tableau Expert | DM me for data consultation!
1yAs someone who just started posting on LinkedIn, thank you. This helps me feel good about where I'm at today and feel better about that 1% engagement I'm seeing. I guess people on LinkedIn hesitate to Like. Weird.
aiXplain Developer Advocate | Podcast Host | Experienced Data Scientist
1yI was literally thinking this morning to ask you some questions on this topic. You saved me several text messages filled with questions to you. Lol. Thanks for sharing this!
lowkey genai engineer
1yThanks for the transparency, I wish you the best for 2024! 🙌
Guest Lecturer @ Stanford University | CEO @ OSS4AI | Chair @ The (AI) Consortium
1ySome people who I want to see year in reviews from: Kristen Kehrer Daliana Liu 🎯 Mark Freeman II Vin Vashishta Jess Ramos, MSBA Megan Lieu Chad Sanderson Jerry Liu