10/09/2025
Is there a right way to post on LinkedIn? Yesterday’s energetic training gave us one answer, research another, and our test last night… well, that raised new questions.
In a LinkedIn training session yesterday, we were told a few things about the algorithms that really made sense, but with due diligence, we couldn’t find the evidence to back them up.
Here’s what the expert explained:
When you post, LinkedIn sends it to a random slice of your network.
If it performs well (scroll-stopping, read, commented on), LinkedIn expands distribution to another random slice.
If it doesn’t, the post quickly loses visibility.
This is why you should think carefully about who you connect with. If your network is full of competitors (not your target audience) or people who don’t engage, your content may never reach the right audience.
This random selection is also why “Tuesdays at 11am” is no longer the golden posting slot. You need to test when your audience is online.
That explanation fits what we see on LinkedIn. But when I looked for hard evidence from LinkedIn or independent studies, I couldn’t confirm the “random slice” part.
So we ran our own experiment last night:
👉 A brand-new LinkedIn account
👉 First ever post
👉 Only 6 connections
👉 One colleague - an MBE with a large, active network - liked and commented
Within 3 hours, that post had over 2,800 impressions. Just under 24 hours later, it's over 22,000 impressions.
This doesn’t prove the “random slice” theory, but it does suggest LinkedIn heavily weights engagement and moves quickly to roll content out to the networks of those who engage.
Which makes it even more valuable to focus on connecting with the right people - the ones most likely to interact with your content and extend its reach.
Here’s what current research does support:
LinkedIn filters posts for quality, then tests them with a small, relevant group (not completely random).
Engagement in the first 1–2 hours is critical, especially comments and dwell time.
Distribution is based on relevance and relationships, not randomness.
Posts can resurface weeks later if someone new engages with them.
So where does that leave us?
The trainer may well be right as their model certainly explains the behaviour we see, but until LinkedIn confirms it, the safest approach is to use it as a working theory and run your own experiments before making big changes to your strategy.
Takeaways I’ll be acting on:
* Build a network of people genuinely interested in your content. Too many competitors or passive connections weaken your reach.
* Test your own timings. Industry data is a starting point, but the best slot is when your audience is active.
* Don’t judge a post on day one. Strong content can resurface weeks later.
What do you make of this?
Have you run any of your own experiments or noticed unusual patterns in how your posts perform? It would be great if we could pool our experiences to help each other get more from LinkedIn.
