Algorithm Insights

It's that time of the year when the legend that is Richard van der Blom publishes his LinkedIn?? algorithm insights report so in this episode I discuss some of the highlights, what we can learn and why we need to treat the results with caution.

Also covered;

• Services page gets a makeover
• Creator mode is going
• Is profile verification enough?

LinkedInformed is a weekly podcast which you can listen to and subscribe to for free on any podcast app. These notes are merely brief overview of what I covered in the podcast so do listen to the full show on Spotify below or simply search for LinkedInformed on your podcast app of choice.

Services.

One of the least impressive, half-cocked features of LinkedIn?? is the services page and it has undergone a refresh, or perhaps better to say that it has mainly changed how it's showcased in your profile...at least for some of us anyway as the change is rolling out.

I have long since given up on this feature. It has had so little attention that I assumed they had given it to an intern to create as project one summer and since forgotten it existed!

On hearing that it had been revamped I visited my services page only to find two previously unseen requests! One was from a 'snake oil salesman' using it to promote a LinkedIn?? automation tool so no big deal but the other was a genuine request sent 6 months ago! To be fair I did find an email alerting me about this (that got lost amongst a myriad of other less important emails from LinkedIn??) but I'm annoyed that there isn't some sort of obvious alert on LinkedIn?? itself, maybe a badge showing the number of unopened requests highlighted somewhere.

As far as I can see this revamp has actually made the feature worse! Instead of your services being highlighted very near the top of your profile (above the fold), they are now being demoted to further down. There is more information visible however, including media such as a video, which can now be native video as well as your most recent review (Premium users only). This would be great if they still had the summary at the top of your profile.

This actually sits beneath your 'About' section which, for creators is quite far down. It remains to be seen where it will sit when creator mode is abolished in March (see below).

Rather than a 'services revamp' I actually think this change has been motivated by a need to create more space at the top of our profiles...for what exactly remains to be seen!

Creator Mode.

2 years ago, LinkedIn?? finally woke up to the fact that the main reason people find this site useful is because of the content that it's users create and so they implemented creator mode to give those authors more tools and support.

I actually thought it was a good move but now they feel it's time to make everyone a creator by dumping the mode altogether.

Here's the announcement...

The interesting bit is the CTA because this suggests that it will now make 'follow' the default CTA for everyone, not just creators. We will have to wait till March to see if this is the case but if so, I'm in favour and see it as a positive move.

Connecting is massively overused in my opinion and should only be utilised in the circumstance where you are committed to building a relationship with someone.

Profile Verification.

I've stated previously that I'm a fan of verification and recommend that everyone do it because we all need to play our part in making LinkedIn?? a safer, more genuine place.

That said, verification alone is not the answer to reducing fake profiles and the prevalence of bad actors. Anyone can still claim that they work for any company they want and that is, and has always been, utterly ridiculous!

This was a highlighted again this week when a CEO of a company got this response from LinkedIn?? when they asked for a fake employee to be removed from their company page

Quote: "You recently notified us about information on a LinkedIn member's profile that you believed was inaccurate. We received a counter-notice from that member and as a result, will not remove the content. If provided with formal judicial process, we'll revisit the matter.Feel free to reach out to us again in the future if anything changes or if you have any further questions."

Thanks, Regards, Simi LinkedIn?? Member Safety and Recovery Consultant

What the actual...!!!!!!!!

So in summary, a bad actor can claim to work for your company and when you try to remove them, if they merely reply to LinkedIn?? with "nope, I don't agree" then you have to go through a legal process (at great expense) to get them removed!

That is nuts!

I've stated many times, the solution is remarkably simple. Employees can only add a current employer if they have an email domain for that company or they are approved by the page admin for that company.

I fail to see why LinkedIn?? refuse to do this but they have stubbornly refused to for many years.

2024 Algorithm Report

This years algorithm report is very very detailed. At 123 pages long it takes some time to get through so I have summarised what i consider to be the main highlights.

Firstly I have two important points to make...

1. The amount of time and effort Richard and his team put into this is quite something. I have a huge amount of admiration for him and for providing free access to everyone.
2. Do not take the results as gospel. These are not rules for using LinkedIn??, the information should be used to guide you to try things out with your own content. The number of posts analysed is extensive, significantly more than you could achieve but it's a relatively moderate number in comparison to all the content published on LinkedIn??. If LinkedIn?? themselves produced such a report (they won't) based on the entire catalogue of content posted on LinkedIn?? it would provide a better analysis but it still wouldn't be right for your audience! This type of analysis will never be better than your own analysis of your posts to your audience. Use this report wisely, it will stimulate ideas for you to try...some will work and others won't.

You can view and download the full report via this post from Richard van der Blom

This years report felt a bit different to me in that the data and analysis is also supplemented with advice. Richard is using the opportunity to guide you on best practice and his advice is mostly good.

• There are more adverts and sponsored company posts than ever before. No surprise really but I still believe that LinkedIn?? are very different from Meta. LinkedIn?? still generate most of their revenue from subscriptions unlike Meta who are purely an advertising company. I'd prefer less ads but I think we are all pretty good at ignoring them these days.
• Organic post reach has dropped for 95% of us. It's still pretty good, just not as good as it has been. This is partly due to the ads but also the amount and quality of content from other authors. The competition for the eyes of your followers has never been greater!
• Saving a post has an impact on a posts reach but also on your 'relevance' to the person who saved it. This makes perfect sense to me and means that we should be guiding our viewers to save our posts. Here's my standard line on all my posts now...

No time to read this now?You can save this for later by clicking on the 3 dots above and right ● ● ●

• Content consumption patterns. When someone engages or saves content, the algorithm is noting what type of content that is (video, text, image etc) and if a clear preference is shown it will adjust their feed accordingly. This seems likely to me so it makes sense to closely observe your key followers and tailor your types of posts accordingly.
• Polls still have some value as a content type. Optimum length is 1 week and 3 answers. Whilst not surprising, this is a neat little tip to remember.
• You can afford to post more text posts more frequently that other types of content. This is possibly more true for short text posts as they take up much less home page 'real estate' although the report doesn't specify that.
• Faux posts such as 'celebrate occasions' do not perform well according to this report. I have found the complete opposite to be true, at least for 'new position' ones. Don't get me wrong, I can't stand these posts but they do seem to perform far better than they should but LinkedIn?? do like to support anything that is job/career related.
• Timing of posts. Richard breaks this down into 3 categories. as seen below;

This one had me confused. The first 60 mins, the golden hour has always been important so I get that. The second one (after 6 hours) influences the next 18 hours which makes sense but what about 2-6 hours? The 3rd one makes no sense to me at all! 'Interactions received after the first day can either prolong or cause it to fade'...erm what? Maybe it's just me but I can't make any sense of that one!

• We've always known about the importance of replying to comments but in this report it is suggested that we must reply immediately. This makes sense but can also be very difficult to achieve, especially when a post does well. My sense is that the marginal gain experienced isn't worth the stress of trying to keep up with all comments on your posts. Do it if you can but don't put yourself under so much pressure, it's not worth the anxiety!
• In several sections actions performed on your own posts are mentioned including commenting within the first hour, then again after 6-8 hours and again after 24 hrs...that all seems like a lot of work to me but it's probably worth a try. Reposting your own post can give a 20% boost to engagement and liking can help but tagging yourself has no impact...and frankly looks weird to me!
• On the topic of tagging, it's well worth reading this section which starts at p48. Mentioning someone in your post is definitely beneficial if they comment (1.5x a normal comment) but be careful of over tagging. Up to4 carefully chosen, active users seems to be the sweetspot
• If you have more than four to tag, Richard recommends tagging the extra ones in the comments where there are no penalties for unresponsive tags. I'm not a fan of this technique because someone who is tagged in the comments is likely to reply to that comment and replies do not carry the post to their followers, which will limit reach.
• Hashtags still have minimal impact - no surprise there!
• Liking has impact but only up to 50, beyond that there is limited extra reach.
• Commenting on other posts after you have posted can make a difference to the reach of your post. Richard suggests adopting a daily habit of 10 long comments per day, presumably on 10 different posts. Short comments are less effective.
• The most interesting page was this P46...

• This is taken from an interview Daniel Roth gave last year. Strictly speaking the first one isn't a distribution mechanism. The alert directs us to a post but that post may not have appeared in our feed. The 3rd one shows the benefit of strategically connecting to targets...albeit a fairly limited one and the 4th demonstrates the longer term benefit of attracting followers who are not connections. This reinforces the benefit of having 'Follow' as the primary CTA on your profile.
• This year there is a comprehensive section on AI starting at P76. The use of AI to create content is resulting in more text only posts and these are typically getting longer. I really like text only posts but I can see that in time, the assumption might be that if it's text, it's AI!

So those are my thoughts. It's a very impressive report and a very useful tool but as previously mentioned, please don't follow it religiously. Try things and judge for yourself whether it's right for your audience.

A big thanks to Richard for once again producing such a comprehensive, thought provoking report.

That's all for this week, I didn't receive any nominations for a post of the week and didn't see any worthy ones myself...oh if only we could filter a post search by engagement levels!

Take care, have a fantastic week and I will be back soon.