FREE TOOL

When should you post on Bluesky?

We analysed 8,373 posts and found the industry's best-time advice does not hold up. The only data that can answer this for you is your own — so this runs the same method against one account.

Public data only. No login, and never an app password — your browser asks Bluesky directly and nothing is stored.

Common questions

What is the best time to post on Bluesky?
For Bluesky as a whole, there probably isn't one. We collected 8,373 English posts from 344 accounts and tested all 31 hour and weekday buckets: exactly one was distinguishable from the baseline, which is about what testing 31 buckets at 95% confidence turns up by chance alone. So the honest reading of our own data is that no universal best time showed up. What can be measured is one account at a time, against its own history, which is what this page does.
Do I need my Bluesky password or an app password?
No, and you should never give either to a page like this. Everything comes from Bluesky's public API, which needs no authentication at all, and your own browser makes the request rather than our servers — so we never see the handle you type. Because it only reads public data, it also works on any public account, not just your own.
Which of my posts does it actually look at?
Up to your 300 most recent original posts. Reposts are skipped because a repost carries someone else's timing, and replies are skipped because they reach a conversation rather than your followers and earn engagement on completely different terms. Anything posted in the last 48 hours is also set aside — likes and reposts are still arriving, so including fresh posts would measure how recent a post is rather than when it went out.
Why does it so rarely say a particular time stands out?
Because it is testing 31 buckets at once, and at 95% confidence you should expect roughly one and a half of them to look special purely by chance. A slot is only reported as a signal when its confidence interval clears your own average, which is a deliberately hard bar precisely so it is not just re-reporting that noise. It also needs at least 20 posts old enough to count before it will draw any conclusion, and at least 5 in a bucket before that bucket is shown at all. Most accounts genuinely do not post often enough to detect a timing effect, and saying so is the point of the page.
Are the times shown in my timezone?
Yes. The buckets are built from your browser's own timezone, so an hour labelled 09:00 is nine in the morning where you are. That is the one place this differs from the study behind it: with hundreds of accounts there was no way to know each author's timezone, so the study had to report UTC and say so. With a single account, your browser already knows.
Will I get the same answer if I run it again?
Yes, for the same posts. The confidence intervals come from a bootstrap — resampling your posts a thousand times — which would normally shift a little on every run. It is seeded from your own sample instead, so reloading the page gives identical intervals. A tool whose confidence numbers wobble each time you refresh teaches you not to trust any of them.

Post at the times you picked, without being there

Set a weekly grid once and PostDaily fills it. Free plan, no credit card.

Keep reading