The LinkedIn golden hour: what actually happens in the first 90 minutes

The LinkedIn golden hour is the claim that engagement in the first 60 to 90 minutes after you publish decides how far your post travels. Half of it holds up. LinkedIn does show a new post to a slice of your network and expand or pull back its reach based on what happens next, so early response genuinely matters. The specific window does not hold up. Buffer, which analysed more than 4.8 million posts, says it directly: there is no official 60 or 90 minute window confirmed by LinkedIn. The number is folklore that got repeated until it sounded like documentation.

What is the LinkedIn golden hour?

Strip out the mysticism and the golden hour describes a plausible sequence. You publish. LinkedIn does not immediately push the post to everyone who follows you, because it has no idea yet whether the post is any good. It shows it to some portion of your network. If those people stop, read, react, comment or share, the post gets shown to more people. If they scroll past, it quietly stops travelling.

That much is uncontroversial and is roughly how every ranked feed works. The golden hour adds two extra claims on top: that the evaluation happens inside a fixed window, and that the window is 60 minutes, or 90, or sometimes 30, depending on which page you landed on. Those extra claims are where the evidence thins out to nothing.

The distinction matters because it changes what you do. If a timer is running, the rational move is to recruit engagement fast, which is how comment pods and “great post!” reply chains got invented. If reach is instead evaluated continuously against relevance, the rational move is to write something worth stopping for and then be present for the conversation it starts. Those are very different Tuesday mornings.

Where does the “first 60 minutes” number actually come from?

We went looking for a primary source and did not find one. Here is what we could trace, in the order of how close each gets to something checkable.

Read that table from the bottom up and the shape of the folklore becomes obvious. The one source that is actually LinkedIn talks about seconds, not minutes, and about whether an individual reader stops scrolling. The sources closest to the platform name no window. The number gets more specific the further you get from anyone who could know it.

None of this makes the golden hour advice useless. It makes it a heuristic with a fake precision bolted on. “Be around after you post” is good advice. “You have 63 minutes” is a made-up deadline, and it produces the worst behaviour on the platform.

Does early engagement actually affect LinkedIn reach?

Yes, and the most interesting evidence is not about hours at all. It is about seconds. LinkedIn's engineering team published its work on using dwell time to rank the feed, and the reasoning is unusually candid. Clicks and viral actions, they note, can be rare, especially for people who read the feed passively, and clicks are noisy indicators of engagement anyway. Dwell time is measurable on every post from every member, whether or not they touch anything.

The team built a model predicting whether a member will skip an update, on the finding that people make a skip-or-not decision within a small number of seconds regardless of what kind of post it is. Adding dwell time features improved model accuracy by as much as 10% across trainings, and the change shipped to all members.

Two things follow. First, the first meaningful judgment on your post is made by an individual reader in about the time it takes to read your opening line, which is why the hook does more work than anything else in a LinkedIn post. Second, plenty of the signal LinkedIn collects is invisible to you. A post can be doing well in dwell terms while your notification tab stays quiet. Judging the first hour by the reaction counter measures the noisiest available proxy.

Which early signal matters most, according to 12,988 posts?

We scored 12,988 English LinkedIn posts by engagement rate and compared the top 10% against the bottom half. We cannot see when engagement arrived, so we cannot test the golden hour directly. What we can see is which kind of engagement is scarce, and scarcity is what a ranking system pays attention to.

Reactions differ by roughly 3x between the groups. Comments differ by more than 14x. That gap is the most useful thing we can tell you about what to chase in the minutes after you publish. A reaction costs a reader one tap and half a second of attention. A comment costs them a sentence they have to compose in public under their real name and their employer's logo. The second thing is much rarer, and it is the thing that moves with quality.

It is also why our engagement rate formula weights a comment as four reactions. That weighting is a modelling choice we made, not something LinkedIn published. AuthoredUp reports, on van der Blom's research, that comments count roughly twice what likes do. Nobody outside LinkedIn knows the real coefficient. What everyone agrees on is the direction, and the size of the gap in our data suggests the direction is steep.

The practical translation: if you have twenty minutes after publishing, spend them producing conversation rather than applause. Full numbers and the rest of the format and hook findings are in our study of 34,000 LinkedIn posts.

What actually counts as engagement in the first hour?

People treat “engagement” as one thing. It is at least seven things, they cost the reader wildly different amounts, and you can only see some of them. Sorting them by what they cost the person doing them is the most useful mental model we have found, because effort is what a ranking system can trust.

Two rows in that table are invisible to you and sit at the top of what LinkedIn actually measures. That is worth sitting with. The first hour you are watching is a partial, lagging view of the first hour LinkedIn is watching. It is also why a post with 40 reactions and one thoughtful comment thread can outperform a post with 200 reactions and silence.

What does a normal first hour look like at your follower count?

Almost nobody publishes a benchmark for this, so expectations get set by screenshots of outliers. Our cohort-wide median engagement rate is 0.40 per 1,000 followers, the 75th percentile is 1.27, and the top-decile cut sits at 5.95. Those are lifetime totals for a post, not first-hour totals, so treat what follows as arithmetic rather than a measurement.

Run the median through a 2,000-follower account and a typical post lands somewhere near a single-digit weighted engagement score over its whole life. A first hour that produces two reactions and no comments is therefore not a failure. It is roughly what the median looks like at that size, and it is why comparing your notifications to a creator with 60,000 followers is a good way to quit. The same arithmetic at 20,000 followers produces a first hour that feels busy without the post being any better.

The number to watch instead is the ratio. Comments per reaction is close to size-independent, and it is the axis on which the top decile of our data separates hardest. If your posts consistently earn conversation rather than taps, the follower count follows. If they earn taps only, more followers just produces more taps.

What should you actually do in the first 90 minutes after posting?

The honest version of golden hour advice is short, and none of it involves recruiting engagement. In rough order of value:

1. Reply to every comment with an actual sentence.Not “thanks!”. A reply that adds something gives the commenter a reason to come back, and a reply thread keeps the conversation alive where other readers can see it. This is the single highest-value thing you can do with the time.
2. Comment on five or six other posts in your feed.You are on the platform anyway. Being visibly active in other people's comment sections is a separate distribution channel that also happens to warm up the people most likely to read yours. We break the arithmetic down in how many comments to leave per day.
3. Answer the direct messages the post triggers. On a post that actually lands, the DMs matter more than the reactions. A reply within the hour reads as attentive. A reply in three days reads as a broadcast account.
4. Put the link in the comments if you need a link at all. Outbound links in the post body are the most consistently reported reach suppressor, and the evidence is genuinely mixed, but a first comment costs you nothing either way.
5. Then close the tab. Refreshing the impression counter is not a growth activity. Set the post down and look at it tomorrow.

Twenty focused minutes covers all of it. If you cannot find twenty minutes in the first 90, publish at a different time rather than publishing and disappearing.

What should you not do in the LinkedIn golden hour?

The fake deadline is what makes people do the following, and the following is what makes LinkedIn feeds worse.

• Engagement pods.A group that agrees to comment on each other's posts on a timer produces exactly the pattern an integrity system is designed to catch: the same accounts, in the same order, with the same empty phrases, every time. Even setting detection aside, pod comments come from people with no interest in your topic, which pollutes the relevance signal you actually want.
• Asking for engagement in the post.“Comment YES if you agree” harvests a word from people who did not read anything. It buys a number and costs you the audience quality that number was supposed to represent.
• Messaging people to ask them to go look. It works once. It costs the relationship the second time.
• Editing the post ten minutes in. There is no published LinkedIn statement that editing suppresses reach, and we could not verify any independent study that measures it, so treat the widespread claim as unproven. The real cost is simpler: a typo bothers you far more than it bothers your reader, and the fix rarely justifies touching a post mid-flight.
• Deleting and reposting because the first hour looked flat. You throw away whatever accumulated reach the post had, including the slow relevance driven distribution that arrives later.

Does the golden hour change how you should time your posts?

It changes one thing. Not the hour that is theoretically best for your audience, but the hour you can actually be present for.

The three largest published timing studies do not agree with each other. Sprout Social, working from nearly 2 billion engagements across 307,000 profiles, puts the strongest LinkedIn windows on Tuesday through Thursday between 11am and 5pm. Buffer, from 4.8 million posts, puts the peak between 3pm and 8pm on weekdays. Hootsuite, working with Critical Truth across more than 1 million posts localized in 118 countries, names 8am to 9am on Tuesday and Wednesday. Nine hours separate the first recommendation from the last. We laid all three out weekday by weekday in the best time to post on LinkedIn.

When three careful analyses of very large samples land that far apart, the sensible read is that the effect they are all measuring is small compared to the differences between their populations. Which frees you to optimise for something you can control. Publish when you have twenty minutes to spend in the comments. A post at a mediocre hour with an author who shows up beats a post at the theoretically perfect hour that nobody, including its author, is around for.

Hootsuite makes the related point that LinkedIn's algorithm rewards content that gets engagement quickly after publishing, and that posting when your audience is offline risks the post never reaching their feed at all. That is the strongest honest version of the timing argument: it is about whether anyone is awake, not about a stopwatch.

What if nobody engages in the first hour?

Nothing expires. The most consequential change to LinkedIn distribution in recent years is the shift toward surfacing posts by relevance rather than recency. Hootsuite reports that a useful post can keep showing up days or even weeks later. Sprout Social describes performance signals being evaluated continuously, with reach expanding or contracting over time rather than being settled in one test. AuthoredUp, citing van der Blom, goes further and reports posts that spark real conversation staying in the feed for two to three weeks.

So a quiet first hour is information, not a verdict. It usually means one of three things, and none of them are fixable by refreshing:

• The hook did not earn the stop. The most common cause by a wide margin. Readers decide in seconds, and if your first line reads like a topic sentence, they never get to the good part. This is a rewrite problem, and the way you open a post is the fix.
• The post asked for nothing. A closed loop gets read and scrolled past. Posts that leave something genuinely open get answered.
• Your audience is not big enough yet for an hour to be a meaningful sample. At 400 followers, an hour might mean nine people saw it. That is noise. See getting your first 1,000 followers for what to do about the denominator.

Judge a post at 48 hours. Judge your writing over 20 posts. One hour tells you almost nothing, which is convenient, because it also means one bad hour costs you almost nothing.

How do you test the golden hour on your own account?

You can run a legitimate version of this experiment, and it is more useful than anything you will read about it, because it measures your audience rather than someone else's.

1. Pick one variable: your presence, not your timing. Publish at the same slot every time. For ten posts, spend twenty minutes replying and commenting immediately after. For the next ten, publish and do not open LinkedIn for six hours.
2. Keep the format constant. Mixing a video into a text-only run contaminates the result. Format differences are large enough to swamp the effect you are testing, as the format comparison data shows.
3. Compare medians, not averages. One post that takes off will drag a mean anywhere you like. Ten posts per arm is the minimum for the median to mean anything.
4. Record impressions and comments separately. Impressions tell you about distribution. Comments tell you about the thing distribution is downstream of.
5. Give it six weeks. Anything shorter and you are reading weather, not climate.

Most people who run this properly find the presence effect is real but smaller than the folklore implies, and that the variance between posts dwarfs it. That is the correct conclusion to reach.

Does posting frequency change the golden hour maths?

It changes the budget. Twenty minutes of comment work per post is trivial once a week and significant five times a week. Buffer's analysis of its own post data found that posting two to five times a week improves both impressions and engagement, which puts most people in a range where the after-post routine has to be sustainable rather than heroic.

The failure mode we see most often is not posting too little. It is a person who posts daily for two weeks with a full engagement ritual attached, burns out on the ritual, and then stops posting entirely. Pick a cadence where you can still do the twenty minutes on a bad day. We walk through the trade-off in how often to post on LinkedIn.

What the golden hour is actually good for

Reframe it and the advice survives. The first 90 minutes are not a window in which an algorithm passes judgment. They are the window in which the people who care most about you are most likely to be reading, because your post is newest in their feed and your notifications are firing. That is the cheapest chance you will get all week to turn a reader into a conversation and a conversation into a relationship.

The algorithm might or might not care that you replied within eleven minutes. The person you replied to definitely does.

The short answer on the LinkedIn golden hour

Early engagement helps. The 60-minute deadline is invented. LinkedIn's own published work points at seconds of attention rather than an hour of activity, the two largest independent studies name no window at all, and one of them says plainly that no such window is confirmed. Use the idea as a prompt to be present after you post, spend the time on replies and other people's comment sections rather than on recruiting reactions, and stop treating a quiet first hour as a death sentence for a post that will still be circulating next week.

One more caveat on our own numbers, since this article leans on them. The cohort skews toward established creators: every account in it had at least 1,000 followers, and the posts come from 65 creators. Engagement counts were captured at scrape time and are lifetime-cumulative, so older posts have had longer to accumulate. The comparisons are correlational. They tell you what winning posts look like, not what will make a post win.