Best Time to Post on Hacker News for a Show HN

Almost everything written about this question is a table of hours, and the tables disagree. This page explains why, shows the one public dataset I could open and check, and then gives you a way to choose a window from your own schedule. If you have not read the rules for what qualifies as a Show HN yet, start with the Show HN checklist and come back.

Why do Hacker News timing studies disagree?

They are answering different questions. Three questions get blurred together:

1. When is the site busiest? That is a question about readers.
2. When does a post have the best chance of reaching the front page? That is a question about competition.
3. When does a post earn the most total attention? That is a question about both at once.

A weekday morning in the United States is the answer to the first question. A Sunday is the answer to the second. Nobody has a clean answer to the third, because the data public analyses use is a count of upvotes, not a count of people who read the page.

The clearest published example is an analysis of the public Hacker News BigQuery dataset by chanind, covering stories from January 2018 to May 2019. The author counted a story as a front page story if it passed roughly 50 upvotes, then calculated, for each hour of the week, what share of submitted stories crossed that line. The finding, in the author's words, is "the opposite answer" from an earlier analysis that treated the busiest hours as the best ones: post on weekends and at times of low activity, to minimise the competition your post faces. The author also adds the caveat that those front page posts likely get fewer views than front page posts at a busier time, so it is a trade.

That caveat is the whole story. Both camps are right about their own question.

Quiet hours or busy hours: what is the trade?

I pulled the hourly table behind that analysis, which the author publishes as a JSON file, and summed it myself. The numbers below are from that dataset only, they are in UTC, and they describe 2018 and 2019. Treat them as the shape of the trade, not as a forecast for next Tuesday.

Window (UTC)	Stories submitted	Reached about 50 upvotes	Share
Every hour of the week	494,674	29,961	6.1%
Monday to Friday, all hours	403,528	22,881	5.7%
Monday to Friday, hours starting 13:00 to 16:00	104,143	5,884	5.6%
Saturday, all hours	46,453	3,393	7.3%
Sunday, all hours	44,693	3,687	8.2%
Sunday, the 6:00 hour	1,161	116	10.0%
Wednesday, the 9:00 hour	3,081	124	4.0%

Read the table with the trade in mind. The weekday 13:00 to 16:00 UTC row is roughly 9am to 1pm US Eastern in summer time, which is when the site has its biggest crowd. It also has the most competition, and the share who cross 50 upvotes is slightly below average. Sunday at 6am UTC, which is the middle of the night on the US East Coast, has the best single-hour share in the data. Notice the denominators, though. The Sunday 6:00 hour holds 1,161 stories across the whole period. A share built on 116 successes is real but noisy, and it says nothing about how many people will read your page.

Two things I will not claim. I cannot tell you today's rates, because the community and the ranking software have both changed since 2019. And I cannot tell you that a post with 50 upvotes got more clicks than one with 30 that landed at a busier hour. The dataset does not contain that.

Why does your availability matter more than the clock?

Because a Show HN is a conversation with a post attached. The official Show HN guidelines say you must have worked on the project personally and be around to discuss it, and they describe the format as something people can try, "give you feedback, and ask questions in the thread." A moderator's published tips for presenting your work add that you should include backstory text, a clear statement of what the thing does, and links to relevant earlier threads.

Now put that next to the ranking rule. The Hacker News FAQ says the basic algorithm "divides points by a power of the time since a story was submitted," with flags, anti-abuse software, discussion demotion, account or site weighting and moderator action also affecting rank. Time is working against every post from the minute it appears. What helps a post early is people reading it, trying the thing and leaving real comments, and what makes people leave real comments is a maker who answers within minutes.

So the failure to avoid is not "posted at a bad hour". It is "posted at a good hour and then went to bed". An unanswered first question looks like an abandoned project, and the thread never recovers its momentum. A mediocre hour with a maker in the thread beats a perfect hour with an empty chair.

Run this planner before you pick a day:

1. Write down three hours, in a row, in which you can answer a comment within about ten minutes. Not hours you could technically be near a laptop. Hours with no meeting, no school run, no flight.
2. Convert the first hour to UTC. Work in UTC for this step only, so nothing changes under you when clocks shift.
3. Match it to the table below.

Your three free hours land on	What that means	What you gain	What you risk
A weekday, 9am to 1pm US Eastern (13:00 to 17:00 UTC in summer time, 14:00 to 18:00 in winter)	The busiest readers are awake	The largest pool of people who might try the app	The most competing posts, and an average chance of crossing 50 upvotes in the dataset above
A Saturday or Sunday, any waking hours	Fewer stories compete	A higher share reached 50 upvotes in 2018 and 2019	Fewer readers on the site; some will be away from their desks
Sunday early morning UTC, which is about 6am in London in winter, late morning in India, early afternoon in Singapore	The best single-hour share in the data	Low competition, and a normal working day for you if you live in Asia-Pacific	Most of the US is asleep, so early replies come from other time zones
Late evening or night where you are, with no hours you can stay up	You cannot be present	Nothing worth the trade	A Show HN with no maker in the thread

If your three free hours match the last row, the answer is to wait a week and change your calendar, not to post anyway.

One note on daylight saving: US clocks change on 1 November 2026, so a "9am Eastern" plan made today moves by an hour in UTC next week. If your window sits across that date, convert after the change, not before.

Does the day of the week matter for Show HN?

Slightly, and the data above shows both directions. In the 2018 and 2019 dataset, Sunday and Saturday had a higher share of stories crossing 50 upvotes than Monday to Friday. Weekdays had far more stories submitted, and more readers. Write it down in the form that helps you decide:

• Weekday: more eyes and more competition. The better choice if you are a US-based maker with a free morning.
• Weekend: fewer eyes and less competition. The better choice if you work a job during the week and the weekend is when you actually have a clear head and a clear calendar.

What does not change with the day is the rule about the product. A Show HN of something people cannot try gets treated as a regular submission or removed, whatever the weekday. And the same rule about the community applies on a Sunday as on a Tuesday: do not ask friends to upvote or comment. The guidelines say so directly for submissions and again on the Show HN page.

What should you have ready before you post?

Everything that can be prepared in advance, so the posting hour is spent on replies. Check these the night before:

• The product works without a signup. The Show HN page asks for ease of trying "ideally without barriers such as signups or emails". If it needs an account, add a demo or a sandbox link.
• The title is plain. Start with "Show HN:" and say what it does. No exclamation points, no praise, no site name in the title. The Hacker News guidelines ask for the original title and no linkbait.
• The backstory text is written by hand. The moderator tips now say to write your text yourself and not to use an LLM to generate or even edit it, because the community is sensitive to it.
• The first comment is drafted if the text box does not show your text, with the backstory, what is different, and the limits you already know about. The structure is the same one I describe for a Product Hunt maker comment, minus the friendliness: on Hacker News, factual and direct reads better.
• Your profile has an email address. The tips say it lets moderators contact you, and sometimes send a repost invite.
• A screenshot and a one-minute demo exist on the page itself, because readers arrive cold and skeptical. Our guide to app screenshots for launch listings covers the sizes.
• The next three hours are clear. Phone on, notifications on, no commute.

If you also plan a Product Hunt launch, keep the two apart. The reasoning for which day to launch on Product Hunt is different, and trying to hit both in the same hour means you are in neither thread.

What do you do if it sinks anyway?

Some good posts sink, and the FAQ explains why you cannot always read the cause: rank depends on flags, software and moderator action as well as votes and time. Here is what the published rules let you do.

• Do not delete and repost. The FAQ and the guidelines both say deletion is for things that should not have been submitted.
• Reposts are allowed, in small numbers. The FAQ says that if a story has not had significant attention in the last year or so, a small number of reposts is fine. The moderator tips say a new Show HN of a release is acceptable only if it is significantly different, that you should link the earlier thread and explain what changed, and that doing it more than once or twice a year is excessive.
• Wait for the invite. The tips say moderators sometimes send a repost invite. You cannot ask for one.
• Read the thread you did get. Even a quiet post may have one comment worth a week of work, such as "I can't tell what this does" or "the signup wall stopped me".

Then move on. A sunk Show HN costs you an afternoon. The slower work of getting your first hundred people to try the app keeps paying whatever happened on Hacker News.

A Show HN lasts a day. A listing stays up. After the thread is over, submit your app to TryMy free: it is reviewed by hand, links straight to your site, and works for free, free-trial and freemium apps.

Frequently asked questions

What time zone should I plan Hacker News timing in?

Plan in the time zone where you will be sitting, and convert to UTC only to compare against data. The public datasets use UTC, so that is the common ground for comparing sources. The readers who matter most for a developer tool are largely in the US, so US Eastern is a sensible reference for a weekday plan. Remember that clocks in the US change on 1 November 2026, which moves a fixed Eastern time by one hour against UTC.

Is Sunday a good day to post a Show HN?

It can be. In the 2018 and 2019 dataset, Sunday had the highest share of stories passing roughly 50 upvotes, at 8.2% against 5.7% for Monday to Friday. The cost is a smaller audience. Sunday works if you will be at your desk on Sunday and want less competition. It does not work if you plan to post and leave.

Can I repost if the timing was wrong?

Not for timing alone. The FAQ asks you not to delete and repost the same story, and says a small number of reposts is fine only if the story has not had significant attention in the last year or so. Moderator tips say a repost of a Show HN should come with a significantly different version, a link to the earlier thread and a note about what changed, and should happen once or twice a year at most.

Does posting time matter for Ask HN?

I found no source that measures it separately, so I will not guess. The same logic applies: an Ask HN exists to get answers, so post when you can read and respond to them. The guidelines also ask you not to post on HN to ask or tell the moderators something, and to send that to hn@ycombinator.com instead.