The best time to post on X (Twitter): what 7,627,422 tweets say
We analyzed 7,627,422 tweets from 14,671 creators. The best times are early morning and late evening US Eastern, and weekends beat weekdays. The biggest lever isn't timing at all.
We analyzed 7,627,422 tweets published by 14,671 creators over the last 12 months. The best time to post on X is early morning US Eastern time (5am to 8am ET) or late evening (9pm to midnight ET), and the best days are Saturday and Sunday. The worst window is the weekday midafternoon. But the honest headline is the size of the effect: the worst posting hour gives a tweet a 23.7% chance of beating its author's usual numbers by 2x, and the best hour gives it 26.2%. Timing is worth about two and a half points. Attaching an image or video is worth almost seven.
Best-time-to-post guides for X are everywhere, but almost all of them average raw engagement across whatever their customers happened to schedule, which mostly measures when big accounts post, not when posting works. We did it differently: every tweet in this study is measured against its own author's baseline, so a 900-follower account's hit counts the same as a 900,000-follower account's. Eden indexes X alongside six other platforms for its Discover research feed, which is the corpus this study runs on. Here is everything we found, including the parts that undercut the premise.
Key findings
- Tweets posted at 5am ET break out 26.2% of the time, the best rate of any hour.
- Tweets posted at 3pm ET break out 23.7% of the time, the worst rate of any hour.
- Sunday is the best day to post on X (26.9% breakout rate); Wednesday and Thursday are the worst (24.1%).
- The entire gap between the best and worst posting hour is about 2.5 percentage points.
- Tweets with an image, video, or GIF attached break out 27.4% of the time versus 20.7% without, a 6.7-point gap that dwarfs every timing effect in this study.
- Long-form posts over 280 characters break out 27.7% of the time, the best of any length band.
- Creators who post daily or more see 25.1% breakout odds per tweet, versus 22.9% for creators posting less than once a week.
A breakout, throughout this study, is a tweet that earns at least 2x its own author's median likes. The full method is below.
How we ran it
- Sample: 7,627,422 original tweets from 14,671 creators tracked in Eden's index, published between August 2025 and August 2026. Original tweets only: no replies, retweets, quote tweets, or X articles. Every tweet was at least 7 days old at measurement time, so like counts had settled.
- Metric: each tweet's like count divided by the same author's own recent median likes. We call a tweet a breakout when it earns at least 2x its author's median. Measuring against each author's own baseline prevents a few huge accounts' habits from becoming "the best time."
- Timezone: all times are US Eastern (ET). X doesn't expose authors' timezones, and no single clock fits a global platform; ET is the least-wrong choice for a US-heavy creator sample. If your audience lives elsewhere, shift the tables to your audience's morning and evening.
- Overall shape: the median tweet in the sample earned 84 likes and the top 10% crossed 1,951. Our index tracks creators worth studying, so those figures describe established accounts, not all of X.
The best hours to post
Breakout rate by publish hour, across all seven days:
| Hour (ET) | Tweets | Breakout rate |
|---|---|---|
| 12am | 206,591 | 25.6% |
| 1am | 208,526 | 25.5% |
| 2am | 223,786 | 25.4% |
| 3am | 233,115 | 25.6% |
| 4am | 250,371 | 25.8% |
| 5am | 268,454 | 26.2% |
| 6am | 291,763 | 26.1% |
| 7am | 332,433 | 26.0% |
| 8am | 392,639 | 25.8% |
| 9am | 439,175 | 25.1% |
| 10am | 454,148 | 24.7% |
| 11am | 456,530 | 24.4% |
| 12pm | 454,344 | 24.0% |
| 1pm | 422,465 | 23.9% |
| 2pm | 397,231 | 23.8% |
| 3pm | 371,142 | 23.7% |
| 4pm | 351,565 | 23.9% |
| 5pm | 328,445 | 24.2% |
| 6pm | 298,134 | 24.1% |
| 7pm | 270,845 | 24.8% |
| 8pm | 260,131 | 25.6% |
| 9pm | 250,281 | 26.0% |
| 10pm | 241,373 | 26.1% |
| 11pm | 223,881 | 26.0% |
The curve has two peaks and one long valley. Breakout odds are best from roughly 4am to 8am ET, dip through the American workday to a 3pm floor, then recover to a second peak from 8pm to midnight. The valley is exactly where publishing volume piles up: the 11am hour carries the most tweets in our sample (456,530) and some of the worst odds, while the 5am hour carries barely more than half that volume and the best odds. The feed rewards showing up when everyone else hasn't.
The practical translation isn't "post at 5am." It's that the crowded American midday, roughly 10am to 6pm ET, is the one window that consistently underperforms. Post before the pileup or after it.
The best days to post
| Day | Tweets | Breakout rate | Median vs own baseline |
|---|---|---|---|
| Monday | 1,105,898 | 25.2% | 0.81 |
| Tuesday | 1,168,143 | 24.2% | 0.78 |
| Wednesday | 1,147,442 | 24.1% | 0.78 |
| Thursday | 1,148,465 | 24.1% | 0.78 |
| Friday | 1,116,912 | 24.4% | 0.78 |
| Saturday | 973,051 | 26.4% | 0.83 |
| Sunday | 967,390 | 26.9% | 0.85 |
Weekends win, clearly. Saturday and Sunday post the two best breakout rates on the two lowest publishing volumes, Monday holds some of the weekend's carryover, and Tuesday through Thursday is the flat bottom of the week. It is the same mechanism as the hourly table: most creators treat posting as a weekday job, so the weekend feed is less crowded, and the tweets that do show up face less competition.
The strongest pockets in the day-by-hour grid (each with tens of thousands of tweets behind it): Sunday 6am to 9am ET at 28.1%, Sunday 3am to 6am at 28.0%, Saturday 6am to 9am at 27.4%, and Saturday 9pm to midnight at 27.3%. The weakest: Wednesday noon to 6pm ET at 22.9%, with Tuesday and Thursday afternoons right behind at about 23.1%.
Does the best time depend on your follower count?
We split the sample by each author's follower count at the moment they posted, then compared the quiet window (midnight to 6am ET) against the rest of the day within each band:
| Followers when posting | Tweets | Breakout rate | Quiet hours (12am to 6am ET) | Rest of the day |
|---|---|---|---|---|
| Under 1,000 | 123,533 | 16.5% | 15.2% | 16.8% |
| 1,000 to 10,000 | 1,104,731 | 22.0% | 23.5% | 21.5% |
| 10,000 to 100,000 | 2,898,278 | 25.1% | 26.2% | 24.8% |
| 100,000+ | 3,502,408 | 26.0% | 26.5% | 26.0% |
Two things stand out. First, the quiet-hours edge is real for accounts over 1,000 followers and biggest in the 1,000 to 10,000 band (23.5% overnight vs 21.5% the rest of the day), then shrinks as accounts grow: past 100,000 followers it is half a point, because a big account's tweets reach its own audience no matter how crowded the feed is. Second, for accounts under 1,000 followers the overnight edge actually reverses. A tiny account's engagement comes almost entirely from the handful of people who know it, so posting when those people are asleep helps nobody. Under 1,000 followers, post when your actual readers are awake and spend your energy on replies, not slot-picking.
Interestingly, this is the opposite pattern from our Substack study, where the quiet-hours edge belonged to small accounts and reversed for big ones. Substack's feed gives small accounts open-feed distribution that X mostly doesn't.
The finding that matters more: timing is small
If you read only one section, read this one. The full spread between the best and worst hour of the week is about 2.5 percentage points of breakout probability. Meanwhile, in the same dataset:
- Media moves the needle almost three times as much. Tweets with an image, video, or GIF attached break out 27.4% of the time; text-only tweets, 20.7%. That is a 6.7-point spread against timing's 2.5.
- Going long beats going short. Tweets under 100 characters break out 24.2% of the time, 100 to 199 characters 24.4%, 200 to 280 characters 24.7%, and long-form posts over 280 characters (a Premium feature) 27.7%. Within the classic 280-character range, length barely matters; the real jump is committing to long-form. How X actually counts those characters is its own topic: see the X character limit, explained.
- Cadence compounds. Daily-or-more posters see 25.1% breakout odds per tweet versus 22.9% for less-than-weekly posters, and they take the bet many more times. There is no penalty for volume in this data.
The one-line summary of this entire study: pick a sensible slot outside the weekday midafternoon, then spend your remaining energy on the tweet, its media, and your cadence, not the clock.
How to act on this
- Schedule into the uncrowded windows: early morning your audience's time, evenings, and weekends. Nobody should be awake at 5am for a 2.5-point edge; schedule your posts and threads instead and let them publish with your laptop closed.
- Attach media. It is the single biggest lever in this dataset, worth nearly three timing-optimizations on its own.
- Post at least daily if you can. Per-tweet odds rise slightly with cadence, and the attempts compound.
- Study your niche's breakouts, not the global averages: the Discover feed surfaces the tweets that beat their author's baseline in your topic, which is this exact analysis running continuously per creator. For sizing up a specific account, here is how to analyze a competitor's X account.
Frequently asked questions
What is the single best time to post on X?
Sunday morning, 6am to 9am US Eastern, if you force us to pick one slot: 28.1% of tweets posted there beat their author's baseline by 2x, against a 22.9% floor on Wednesday afternoons. The broader rule is early morning or late evening ET, any day, and weekends over weekdays.
What is the worst time to post on X?
The weekday midafternoon. Wednesday from noon to 6pm ET is the weakest pocket in our grid, and 1pm to 4pm ET underperforms on every weekday. It is also when the most tweets are published, which is probably not a coincidence.
Does the best time to post change with follower count?
Yes. The early-morning and late-evening edge is real above 1,000 followers and strongest between 1,000 and 10,000 (23.5% breakout overnight vs 21.5% the rest of the day). Under 1,000 followers it reverses: your engagement comes from people who already know you, so post when they're awake.
What about Threads?
Our Threads index is still too small for a study we would stand behind (14,386 posts from 605 creators over the same year), so we're not publishing tables for it yet. Directionally it rhymes with X: Sunday was the best day in that sample too. We'll publish the full cut when the sample earns it.
Does posting through a scheduler hurt reach on X?
We have no evidence it does. Scheduled posts in this dataset publish through the same X API paths as native scheduling, and the breakout math measures each account against itself, which would surface a systematic penalty. The mechanics are in how to schedule a thread on X.
Where does this data come from?
From Eden's own social index, the same corpus that powers its research features across X, Threads, LinkedIn, Instagram, TikTok, YouTube, and Substack. For this study we took every indexed original tweet published in the 12 months to August 2026 that was at least a week old, 7,627,422 tweets from 14,671 creators, and measured each against its own author's median likes. No third-party estimates, no surveys, no vendor data.
We built this study because every existing best-time-to-post guide for X measures when tweets get raw engagement, which mostly measures when big accounts post. If you want the same research running on your niche daily, plus scheduling for posts and real native threads, Eden's free plan includes 100 saved items and 50 AI credits with no card required. Start free.