Best Time to Post YouTube Shorts: What Our Data Shows

Search for the best time to post Shorts and the top pages disagree with each other. Some say Tuesday, others Friday. Some say late morning, others late afternoon or the middle of the night. Most do not say how they measured it. This guide takes a different route: we tested it on real Shorts, and we report what the test can and cannot show. It builds on our general guide to the best time to post on YouTube, which covers YouTube's own statement and how to read a channel's history.

What does YouTube say about when to post Shorts?

YouTube Help gives no separate rule for Shorts. Its performance FAQ says publish time is not known to impact a video's long-term performance, and that the recommendation system aims to deliver the right videos to the right viewers regardless of when a video was uploaded. It says publish time is important for formats such as Live and Premiere. It adds that publishing when your audience is most active can be beneficial for early viewership, but is not known to impact long-term viewership.

The tool YouTube points to is the "When your viewers are on YouTube" report in YouTube Analytics. YouTube Help describes it as showing your audience's online activity across your channel and all of YouTube, including the formats viewers watch across videos, live streams and Shorts, and says it can help you decide what time and day to upload.

Put together, that leaves a narrow question. If timing helps at all, it may help the first hours. Whether it changes how a Short does over weeks is what we set out to check.

How did we test it?

We started from the 115 channels we drew at random from the ones our site tracks, which lean toward popular or looked-up channels (median 274,000 subscribers), and read each channel's latest 50 uploads through the YouTube Data API on September 25, 2026. The API gives the time a video was published to YouTube and its public view count, and no flag for Shorts. So on September 26 we asked YouTube whether each video of 3 minutes or less is served at its youtube.com/shorts/ address. That is behavior we observed, not a documented feature. It found 3,058 Shorts, and our Shorts length and size guide has the details.

We kept Shorts that were at least 14 days old, had views, and came from a channel with at least 8 such Shorts. That left 1,710 Shorts from 64 channels. The steps that follow are our own arithmetic:

• For each Short we compared its views with the same channel's other Shorts at the same age, on a log scale. That way a channel with big audiences is compared with itself, and an older Short is not favored over a newer one. Among Shorts this old, doubling the age added only about 7% to views on average.
• We grouped Shorts by the UTC hour and weekday they were published, and averaged the comparison in each group.
• To test whether any group difference was more than luck, we shuffled the hour labels among each channel's own Shorts 5,000 times and counted how often a random shuffle produced a difference as large as the real one. That share is the p-value. A small one, for example under 0.05, would say the pattern is unlikely to be chance. A large one says it looks like noise.

What did the data show?

The 8 blocks span a range from 0.92 to 1.12 times, and the shuffle test puts the chance of seeing a spread that large from random labels at p = 0.77. The block that looks best, 21:00 to 23:59 UTC, has one of the smaller samples (126 Shorts) and does not stand out from chance.

Days do no better: p = 0.22 for the seven weekdays, and p = 0.50 for weekend against weekday. Friday is lowest and Monday and Sunday highest, but each is within what shuffled labels produce.

We also asked whether local time changes the answer. For the 32 channels in countries with a single time zone (788 Shorts), we converted upload times to the channel's local time and used four blocks of the day. Local time of day gave p = 0.70 and local weekday gave p = 0.25. The country is where the channel is based, not where its viewers are, so this is only a rough check.

There is one faint signal, and it is worth describing plainly. When we split the day into 24 separate hours, the difference between hours is slightly larger than chance would give (p = 0.03). Hours explain about 2.0% of the variation between Shorts, against about 1.2% expected from random labels. It does not amount to a pattern you could use:

• No single hour stands out once you allow for having looked at 24 of them (p = 0.33 for the best hour).
• Neighboring hours flip sign. In our data 10:00 UTC came in at 0.86 times, 11:00 at 1.38 times, 12:00 at 1.00 and 13:00 at 0.73, which is what noise in small groups looks like.
• Splitting the channels into two random halves, the hour-by-hour pattern of one half matched the other only weakly (average correlation 0.15).
• We ran several tests, and with that many, one p-value near 0.03 is not unusual.

How big is the timing effect compared with everything else?

Small. The middle 80% of the Shorts in our sample ran from 0.28 times to 4.25 times their channel's typical Short of the same age. A Short from the same channel can do three or four times better or worse than its neighbor, and the biggest gap between two 3-hour blocks was about 22%. Grouped into blocks, the hour of upload accounts for well under 1% of the variation in views. Whatever else differs between two Shorts from the same channel, such as the topic or the opening seconds, has far more room to matter than the hour does.

What can this test not tell you?

• Upload time is not the moment views started. A Short can sit for hours before the feed picks it up.
• We do not know where viewers were. The audience's time zone is the one that matters, and our channel-country check is a rough stand-in.
• Our channels lean large, with a median of 274,000 subscribers. A channel with a few hundred subscribers may behave differently.
• The view counts are public counts, taken on one day. YouTube Help says a Short view counts when a Short starts to play or replays, with no minimum watch time.
• We only see 14-day-old Shorts, so a timing effect on the first hours would not show up. YouTube Help says exactly that kind of effect is possible.
• Channels are habitual. The median channel in our sample published 53% of its Shorts inside one 3-hour window, so we could only compare hours a channel already uses.
• It is an observational test. It shows what went together, not what caused it.

When should you post your Shorts?

1. 01

   Open your own audience report

   In YouTube Analytics, open the report called "When your viewers are on YouTube". It is your audience, in your time zone, and it includes Shorts.

2. 02

   Pick a schedule you can keep

   YouTube Help does not say a fixed time changes long-term performance. A rhythm you can hold for months is worth more than a perfect hour you drop after two weeks.

3. 03

   Change one thing at a time

   Move your usual slot for a run of Shorts and keep the topic and style as they are. Our suggestion is at least 10 Shorts on each side of a change.

4. 04

   Compare medians at the same age

   Compare the median views of the new group with your earlier group when both are the same number of days old. One Short that does well says little.

5. 05

   Look at your own history first

   Our Best Time to Post tool reads a channel's 50 most recent public uploads and groups them by UTC day and hour. It does not separate Shorts from long-form videos, and the Channel Audit shows how each upload compares with your median.

FAQ

Related reading