X (Twitter) Best Time to Tweet Calculator

1. Sort your past posts into two to four time slots, such as weekday morning and weekend. Use one time zone throughout.
2. Open the analytics for each post. Add up the impressions and engagements for each slot and enter them with the number of posts.
3. Pick what to rank by: impressions (Impr.), engagements (Eng.) or engagement rate. Then read the ranked table, the lift and the sample warning.

What each result tells you:

The best time to tweet is the posting slot in which your own posts have earned the most impressions or engagements. It is a property of your audience, not a universal hour.

X publishes no best time to post and no hourly activity chart. Lists of "best times" come from third parties studying other accounts, in other time zones, with other followers.

X Help does describe how posts reach people. The Following timeline displays posts from only the accounts a person follows. The For you timeline adds recommended posts, picked with a variety of signals such as popularity and how people in their network interact. Followers who are active when you post are the first people able to interact with it.

This tool is for creators, brands and community managers who want evidence from their own history. It ranks what happened. It does not promise the next post will repeat it.

The calculator turns each slot into per-post averages, pools all posts into one baseline, and ranks the slots against it.

1. Impressions and engagements are each divided by the posts in that slot. Each result is a mean.
2. The baseline pools every slot: all impressions divided by all posts, and the same for engagements.
3. Slots are sorted from highest to lowest on the metric you picked.
4. Lift compares each slot with the baseline. A slot at 3,200 impressions against a baseline of 2,735.3 has a lift of +17.0%.

The X developer documentation defines impressions as the count of times a post appeared on a user's screen, and it lists engagements as a separate metric visible to the account owner.

Example values only, not benchmarks. A creator reviews 34 posts from the last two months.

Ranked by impressions, weekday morning wins with a lift of +17.0% and a lead of +14.3% over weekday evening. Ranked by engagements, weekday evening wins with +22.1%. Meaning: mornings brought the most eyes, and evenings brought the most interaction. The better slot depends on the goal.

Posting time is one of several things that differ between two groups of posts. These factors shape how much the ranking means.

Number of Posts per Slot

One post that takes off can lift a whole slot. The tool flags any slot with fewer than 5 posts. That flag is our caution rule, not an X rule.

Where Your Followers Live

Followers spread across time zones are active at different hours. Your 9 AM is evening somewhere else, which is one reason no single best time exists.

What You Posted

Topics, media and wording differ between posts, and they change results too. Compare posts of the same kind.

Recommendations

A post recommended in For you timelines can gain impressions long after it was published. One such post can make a slot look strong for reasons unrelated to the clock.

When You Record the Numbers

Impressions keep rising for a while. Record each post after the same wait, such as 7 days.

The three metrics measure different goals, so the same slots can rank in a different order.

Lift is a percentage difference from your baseline. The percentage increase calculator shows the same sum for any two numbers.

Planning a Regular Schedule

Test two slots for a month each and keep the better one. X Help describes scheduling a post for a set date and time from the Composer in an X Ads account.

Checking a Published "Best Time"

Enter your posts from the suggested hour as one slot and your usual hour as another. Your data settles it.

Separating Timing from Frequency

How often you post is a different question. Compare two posting rates with the tweet frequency impact calculator.

Comparing Platforms

The same method works on other networks. See the Reddit post timing calculator and the Instagram posting time ROI calculator.

1. Trusting a Generic Best Time

Claims such as "post at 9 AM on Wednesday" have no source in X documentation. The old version of this page made such claims from fixed tables, and we removed them.

2. Believing an Hourly Heatmap

The old heatmap multiplied four made-up tables: audience activity, content type, goal and day of week. It showed no real data.

3. Comparing Totals

A slot with more posts has more total impressions. Always compare per post.

4. Mixing Time Zones

Label every slot in one time zone. Mixed labels put the same hour into two slots.

5. Reading a Small Gap as a Win

In the second preset, 3,000 against 2,900 impressions per post is a gap of 3.4%. A few more posts can reverse it.

6. Reposting the Same Post at Several Times

The X authenticity policy prohibits repeatedly posting identical or nearly identical posts. Test with new posts.

The averages are exact for the numbers you enter. A ranking of past posts cannot prove that timing caused the difference.

X publishes no best time. The best slot differs by account and audience, so the reliable answer comes from comparing your own posts, which this calculator does.

It can. Followers who are active when you post are the first able to see and interact with it. X Help also says the For you timeline adds recommended posts using a variety of signals, so timing is one factor among several.

More posts give a steadier average. The tool marks any slot with fewer than 5 posts as provisional. That flag is a caution rule, not a guarantee.

Rank by impressions for awareness and by engagements for interaction. In the example on this page the two pick different slots: weekday morning and weekday evening.

Use one time zone for every slot, normally your own. If most followers live elsewhere, label slots in their time zone and keep it fixed.

Yes. X Help describes scheduling a post for a specific date and time from the Composer in an X Ads account, and says the post goes live with the timestamp you set.

No official source for them exists. The old page printed hours from hard-coded tables with no data behind them, so they were removed.

Not necessarily. Run the comparison separately for different kinds of post, such as replies and original posts, if you have enough of each.

Repeat it when your audience or content changes, or every few months. A ranking from old posts describes an older audience.

Anyone with only a handful of posts per slot. With so few posts, the ranking reflects single posts more than posting time.

No. The calculation runs in your browser and nothing is sent to our servers. Anything you choose to Save stays in this browser only.

• About your For you timeline: followed accounts and recommendations (X Help Center, read 2026-10-03; opens in a browser, blocks automated checks).
• X API metrics: impressions, engagements and the 30-day limit (X developer documentation, read 2026-10-03).
• Scheduled Posts in X Ads (X Help Center, read 2026-10-03; opens in a browser, blocks automated checks).
• Authenticity policy: engagement spam and follower inflation (X Help Center, read 2026-10-03; opens in a browser, blocks automated checks).
• Measures of location: definition of the mean (NIST/SEMATECH e-Handbook of Statistical Methods).