My friend thought he had hacked the LinkedIn algorithm. He wrote about the Lazada retrenchment, and his post impressions spiked. We both assumed the event had carried the post. So, when GovTech announced retrenchments, he tried the same thing again. This time, nothing happened.
That is the problem with trying to crack the LinkedIn algorithm: we extrapolate one good result into a rule, but it does not hold up for the next post. I do not proclaim to know the algorithm, but I can show you what happened on my account, including the posts that failed.
LinkedIn metrics mislead when treated as a matrix: use the 90-Second LinkedIn Post Test instead
The 90-Second LinkedIn Post Test. Eileen’s test asks whether a post is interesting, useful or recognisable before its likes, comments or format are treated as evidence of success. You cannot reduce LinkedIn reach to one hack, format, posting time or frequency. LinkedIn’s 2026 feed-ranking system may help relevant posts reach people beyond your immediate network, but individual results are uneven. In my experience, the stronger posts are interesting, useful or recognisable; saves and lasting reach tell me more than a spike in likes. Use timing and format as experiments, not rules.
LinkedIn's 2026 algorithm rewards relevance and sustained attention
It is more nuanced than some LinkedIn influencers suggest
In March 2026, LinkedIn said it was rolling out a feed-ranking system. Instead of just pushing posts to your first-degree connections, it would push posts to people outside your immediate network, provided they are interested in that area. That is good news for people with great content but small reach: it can help them get discovered. Read LinkedIn’s March 2026 feed explanation for details, but be warned: it is dense.
Coincidentally, my impressions also started increasing around the same time, so I could have been a lucky beneficiary of the new algorithm. My impressions dipped in July because I decided to shift away from writing about personal problems, such as job seeking, to institutional problems, such as branding and communications.
Saves and lasting reach matter more than a spike in likes
Likes can be vanity metrics; saves can reveal lasting usefulness
My friend once told me that his posts with the most likes were the most superficial: posts that took him five minutes to write and showed him collecting an award, announcing a promotion or speaking at an event. He sighed that the book review which took an hour to write—something he thought his whole industry should read—drew just thirty likes.
So why am I asking you to write thoughtful posts if shallow posts “perform better”? It only seems that way because we are looking at the wrong metrics.
Take this post I wrote a year ago, in the left column of the diagram below. It is the typical “I’m honoured to have spoken at this conference” post. It raked up 95 reactions, as much as another long-form article I wrote about a friend who built a multi-income, multi-portfolio career. But the latter post raked up 33 saves too, meaning readers found it useful enough to bookmark and return to. The second post has greater longevity, as LinkedIn is more likely to keep promoting it, helping more people know about me.
I know that LinkedIn posts can keep circulating. Just last week, someone connected with me because of a post I had written nine months earlier. LinkedIn was still circulating it.
That is what a good LinkedIn post should do: bring you into rooms that you otherwise would not be able to enter, even long after you have forgotten about the post.
The best LinkedIn format depends on the story and audience
Articles suit the stories I want to tell
I prefer writing articles because they are the only format that can accommodate proper stories.
But I have heard people ask whether articles put readers off because they need one extra click to go into the article, compared with posts. Not many people know that they can click on the picture to read the article, so they often just read the post and wonder why I stopped my stories halfway.
My carousel experiment performed poorly, but your audience may differ
I converted some of my older articles into carousels. The results surprised me: both carousels performed poorly, getting half as many reactions as the original article. But there are confounding factors. The main one is that I have curated an audience used to long-form arguments. Someone with a different audience may see the opposite result.
Video did not work for me
There are claims that LinkedIn videos do particularly well. LinkedIn pushes them out aggressively in a bid to compete with TikTok and Instagram. Mine performed so poorly—I think it got six likes—that I took it down after half a day.
Post ideas that are interesting, useful or recognisable
Aim to make readers think: “That is interesting”, “that is useful” or “that is me”
Across my best-performing posts, I keep seeing the same three ingredients. Not every post needs all three in equal measure. But if it is not interesting, useful or recognisable, no posting time will save it.
1. Can you make readers think, “That is interesting”?
The opening contains a contradiction, an irony or an unfinished question.
For example: “You just sabotaged your best-performing individual contributor.” Promoting someone is supposed to reward them. How could it backfire? That gap pulls the reader into the story.
2. Can you make readers think, “That is useful”?
The post helps the reader solve a perennial problem with a simple solution.
In “Why your top talent quits without warning, and the $10 fix,” readers find out that a small Starbucks card can surface a problem before a good employee resigns.
3. Can you make readers think, “That is me”?
The reader recognises their own frustration, ambition or private thought in the story.
“Let’s take this offline” worked because people have strong feelings about that line but often do not say them out loud. The post became somewhere they could vent, disagree and add their own stories.
You can learn how to evoke these three elements in this article. For a complementary Block 173 guide, read How to write a LinkedIn post with substance?
Test posting times instead of treating one time as a rule
My starting point is 7:15 am on Monday, Tuesday or Wednesday—not a magic time
That is when people are commuting to work, and it gives the post a full working day to travel.
Then I accidentally broke my own rule. I was overseas and scheduled a post for 7:15 local time instead of 7:15 Singapore time. It went out at 10:15 am Singapore time and became my best post ever.
I would not conclude that 10:15 am is the new magic time. The post, “Why your top talent quits without warning?”, was gossipy and gave people social cover to say things they otherwise would not.
Weekends and public holidays may still be worth testing
I used to avoid posting on weekends and public holidays because I assumed fewer people were thinking about work. But I am less certain now. LinkedIn has become Facebook for adults under the cover of professional development. People announce pregnancies, proposals and birthdays there.
So I posted during the first week of Chinese New Year and received decent reach. After all, what else is there to do while visiting relatives?
A sustainable weekly rhythm protects quality
Once a week lets me protect quality
I normally publish once a week because a good article takes me hours to conceive and write.
But I experimented with daily posting from 27 to 31 July. To be fair, I was not writing five new articles. I was mostly turning older pieces into carousels. But I did not see any spike in my profile views or impressions.
There are many reasons for that, including benign ones: my audience may not have wanted to reread older pieces they had already seen, or they may simply not be tuned in to long-form articles rather than carousels.
Even if daily posting worked, I would not choose it
The daily pace made it hard for me to maintain quality. Even a reader who likes you can become tired of seeing you every day—I know because I have unfollowed people for exactly that reason. Third, I burnt out. Every night became a struggle.
I would rather publish one strong article a week than five shallow posts.
Collaborations and established platforms can extend reach
Posting more is not the only answer. There are two other ways to get a good idea moving.
1. Borrow reach through a collaboration
Work with a writer whose audience fits the story you want to tell. I recently wrote what I jokingly call two infomercials for freelance friends: one on building a portfolio career and another for a resume writer.
They did not feel like advertisements because the reader received a story and a lesson, and the person or service merely featured as part of that storytelling.
2. Borrow a platform
If you have a useful idea but no audience, you can consider writing for Block 173.
It is named after the block where I grew up in Singapore. For non-Singaporeans, a void deck is the open ground floor of a public-housing block: a third space where neighbours gather to chat.
My grandmother found my first tuition teacher after hearing that the son on the 16th floor had scored 270 for his PSLE after going to this teacher. She also learnt how to make my favourite achar, a spicy pickle, from an auntie she met downstairs.
Block 173 brings that idea online. For Singaporeans, I want it to become the career equivalent of Miss Tam Chiak. When you cannot decide where to eat, you look there. When you do not know what to do next at work, I want you to look there, because you will be able to find a real person who has done it and see how they did it.
That means relying less on anonymous hearsay on Reddit. You can read interviews about moving into a portfolio career, deciphering a headhunter’s signals, changing industries without direct experience, writing a resume or building a reputation on LinkedIn.
If you have a useful story but not the audience for it, send it to me at the Letter Box. If your company wants people to understand its work without making them read an advertisement, drop me a note too and I might be able to help.
Build on the distribution advice by learning how to write a LinkedIn post with substance and make your perspective useful to senior people.
Frequently asked questions
Can you really crack the LinkedIn algorithm?
Not with one repeatable trick. A post that benefited from one news event did not work the same way with another, and I do not claim to know the algorithm.
Are LinkedIn articles, carousels or video best?
There is no universal winner. I prefer articles for proper stories; my carousels got half as many reactions as the original articles, but a different audience may see the opposite result. My own video did poorly.
When is the best time to post on LinkedIn?
My preferred starting point is 7:15 am on Monday, Tuesday or Wednesday, but a 10:15 am Singapore-time post became my best ever. Treat timing as an experiment, not a magic answer.
Should you post on LinkedIn every day?
Not necessarily. My daily-posting experiment did not lift profile views or impressions, and the pace made it harder to maintain quality and avoid burnout. I would rather publish one strong article a week.
About Eileen Soh
Eileen Soh is a Singapore-based LinkedIn writer and thought leadership advisor
She teaches professionals how to catch a senior decision-maker’s attention in 90 seconds and give them a reason to remember you.
Before she started writing publicly, Eileen spent more than a decade writing for senior decision-makers in government and Fortune 500 companies. Working for readers with little time taught her to get to the point quickly, make every line worth reading and show immediately why something matters. That discipline still shapes her writing and teaching today.
Today, she teaches the same discipline. She conducts workshops on how to use those 90 seconds to catch attention on LinkedIn and in networking, and build a professional reputation that opens doors.
She has spoken at Women in Tech Singapore, RHT Law, the Nanyang Academy of Fine Arts (NAFA), LASALLE College of the Arts and the Singapore University of Social Sciences (SUSS). She has also been featured in The Straits Times and on The Flying Dutchman’s Morning Drive radio show.
Leave a Reply