Can You Game The LinkedIn Publishing Algorithm?

a-post

5 surprising factors that don’t seem to affect published post engagement on LinkedIn and 5 not so surprising ones that do.

I have been publishing on LinkedIn for eighteen months now, and have written a hundred or so posts. I was looking at some of my posts and the engagement they had generated and wondered if any of the obvious factors really had any bearing. So I compared how my posts had done with other people who publish on LinkedIn. I compared their statistics with mine to see if there were clues that were precursors to success. I came to the conclusion that the following have little or no bearing on how much engagement a post gets.

  1. The total number of posts published on LinkedIn

Your total number of posts published seems to have no bearing on the success of future posts, that is people who have many more posts than I do don’t seem to average more views and engagement with their posts than I do. Of course their aggregate views may be higher.  If I have 20 posts at 200 views and they have 40 posts at 200 views, they have more total views.  

Lesson learned:

  • Don’t be deterred from starting to publish your own posts on LinkedIn  because you can never “catch up” to someone who has posted every week for the last two years. You are starting with a fresh page, but so is that prolific person who posts each week.

      2. Engaging with other people’s posts and updates? No discernible effect.

I had a look through the activity for ten people who publish their own posts on LinkedIn. More engagement with other people’s posts and status updates does not necessarily appear to lead those people to your posts. Several people I looked at have three and four times the number of status updates and comments on other people’s posts as I did in the last month, yet seemed to average the same post statistics as I did.

Lesson learned:

  • You don’t need to strive to be visible on LinkedIn in order for your posts to be successful

      3. Monster posts don’t have much, if any, spillover effect on future posts.

Having a monster post with a lot of  views and engagement doesn’t necessarily help with the next post’s reception. I have even seen this with the influencer posts

Jeff Haden November 17, 2015 post: 546,000 views

Jeff Haden’s next post on December 3rd, 2015:  23,000 views  

Lesson learned:

  • Don’t think that a monster post is necessary to be successful.
  • And if you have a monster post, don’t think you can cruise from now on based on that post. You are guaranteed no views or comments on your next post. Just ask Jeff.

     4. Post frequency has no impact  

Publishing more often doesn’t seem to change the results on a per post basis. No difference. Once a month, twice a month, every week, multiple times a day (!!). I reviewed several people who post a lot less often than I do and much more often than I do and our statistics were very close.

Lesson learned:

  • Post on your own schedule, not someone else’s.

     5. Do people with larger numbers of followers get more views per follower? It doesn’t seem so.  

I have 5500 followers. Jeff Haden has 890,000 (between the time I wrote this and published it he may have picked up another 5500).  I did some calculations based on the last nine posts we have each published (Jeff has only published nine times in the past year). Some interesting results:

Per post

              Jeff averaged:                                                        I averaged:

              93,000 views                                                         411 views

              1400 likes                                                               47 likes

              300 comments                                                      12 comments

Lessons learned:

Jeff’s number of followers is more than 160x mine. So three observations.

  • Jeff has better statistics than I do (that was the easy observation)
  • Mine might be comparable to Jeff’s if I took the time to increase the size of my network 160x
  • Jeff’s views and likes per follower are slightly higher than mine, but his comments are much lower. This argues that there is no exponential engagement accelerator with a larger following (for example a following twice as big doesn’t yield two and a half times the engagement).   

Grand conclusion:

You can’t game the LinkedIn algorithm with any of these things.

However, based on my own experience, here are five factors that do seem to affect overall engagement:

1) Writing quality. Not as big a factor as you would think, but being able to write in an interesting and engaging style helps.

2) The photo or illustration you use. Having no photo is bad. Using a stock photo is better than nothing. Using something original beats using a stock photo. How much stock and original photos are is open to debate.  

3) Headline. Better headlines lead to more views and engagement. I know this from my own experience.  

4) The topic of your post. On LinkedIn, a post on social selling will do better than a post on coal mining in Silesia (now I have probably offended all the Silesians)

5) Luck. LinkedIn has decided to put my posts in Pulse channels a few times. I usually tweet at them to feature my post (except when I have been skeptical of something LinkedIn has or has not done) and have been successful maybe one in ten times. Considering LinkedIn has that 130,000 plus posts to choose from every week, I consider myself lucky to have been featured at all.  

And one last one that qualifies as neutral:

Building up your own following

Regular readers who like, comment and share your posts – should help. I have a pretty good following (thank you all), but ironically, due to the nature of LinkedIn’s lousy notifications algorithm, hardly any of them actually do get notified when I post.

Hmm…looks like I won’t be notifying LinkedIn about this post either.