Shorts beat TikTok 11.2k to 10.4k. Most people never post there.

Most people who make short videos for a business treat YouTube Shorts as an afterthought. TikTok is where the reach is, Instagram is where the customers are, and Shorts is the place you upload to if there is time left over. In our own 100-reel experiment, there was one week where the same reels got 11.2k views on Shorts and 10.4k on TikTok. The afterthought won.

That raises a question most short video advice skips: if the reel already exists, why is it not on every platform that will show it to strangers, and what does it take to automate that without turning it into a chore?

The experiment and what Shorts did with it

The setup: one new account per platform, zero followers, 100 reels over 30 days, every reel written, voiced, edited and scheduled by the pipeline that became Ultim. No ads, no cross-promotion, no existing audience.

In our own 100-reel experiment, TikTok gave a median of 303 views per reel, a best of 1,728, and 48,254 views in total across the 100 reels. Instagram with zero followers was close to dead at about 800 views across 20 reels. YouTube Shorts sat between the two most weeks and ahead of TikTok in at least one (11.2k against 10.4k).

The reels were byte-for-byte the same. The difference was entirely in how each platform handles a new account. TikTok and Shorts both test new content on non-followers. Instagram, for a brand new account, mostly does not. Shorts is therefore not a secondary platform for a business starting from zero. It is one of the two primary ones.

Why this surprised us

The common assumption is that Shorts is a mirror of TikTok with a smaller audience for short content. In practice it behaved like an independent test with an independent pool. A reel that got a mediocre pool on TikTok could get a good one on Shorts, and vice versa. Posting to both was not duplication. It was a second roll of the dice with the same reel.

What travels between platforms and what does not

Based on our analysis of 4.5 billion TikTok posts (2014 to 2026), the features that separate the top 1% of posts from the rest are length, caption, hook type, sound and posting hour. These are properties of the reel and the upload, not of the platform, which is why the same creative choices apply to Shorts.

The levers, briefly:

• Length: 61 to 120 seconds carries a lift of 3.08; 8 to 15 seconds carries 0.63; photo posts carry 0.24.
• Caption: 21 to 100 characters with no hashtags, 1.7 to 1.8; empty caption, 0.27; 1 to 2 emoji, 1.53.
• Sound: own voice or own recording, 1.09; someone else's music, 0.79.
• Hooks: GRWM, vlog and ASMR 3.32; "part N" series 2.90; how-to 2.65; tips 2.49; question 2.08; storytime 1.81; POV 1.39.
• Posting hour: 10:00 to 18:00 local, 1.04 to 1.12 with a peak at 15:00 to 16:00; 21:00 to 02:00, 0.78 to 0.84.

The 61 to 120 second window is particularly relevant for Shorts, where longer short-form has a natural home. A reel scripted to 75 seconds with a story arc is a native fit on both platforms. A 12 second clip is the weakest format on both.

The series effect

One trend in the hook data matters more for Shorts than anywhere else. Most hook types lose 10 to 15% of their lift every year. The only one whose lift grows year over year is the "part N" series, from 2.81 in 2024 to 3.01 in 2026. Shorts lives next to a subscription button and a channel page where parts sit in order. A numbered series is the format that turns a Shorts viewer into a channel subscriber, and the data says it is also the format that is still gaining on TikTok.

What YouTube Shorts automation has to get right

Automation for Shorts fails in predictable ways. Here is the checklist we built Ultim around.

Upload as a Short, not as a video

A vertical file under the Shorts length limit uploaded through the right path is treated as a Short. The same file uploaded as a regular video is a tiny landscape-letterboxed video nobody will find. Any automation must use the Shorts path explicitly.

Title and description are the caption

On TikTok the caption is one field. On Shorts the title does the job of the caption and the description is mostly for search. The caption rule from the data still applies: one plain sentence of 21 to 100 characters, no hashtag wall. Ultim generates a title in that window and keeps the description short.

Own audio, not a library track

A trending song under a Short is someone else's music (0.79). An AI voiceover recorded for the reel is original audio (1.09), and it does not expose the business to the licensing restrictions that come with library music on a business channel. This is one of the quiet reasons Ultim records a fresh ElevenLabs voice track for every script rather than mixing a soundtrack.

Post in the afternoon, local time

The posting hour lift (10:00 to 18:00, peak 15:00 to 16:00) comes from TikTok data, but there is no reason to expect Shorts viewers to be awake at different hours. A scheduler that posts at 15:30 in the business's local time is doing the only timing work there is to do. Day of week barely matters (0.90 to 1.05), Sunday weakest.

Measure at 24 and 48 hours, per platform

In our own 100-reel experiment, TikTok gave one pool of views in the first 12 to 24 hours and then went flat. Shorts behaved similarly. That means the 24 hour number is a reliable early read and the 48 hour number is close to final. Ultim records both for every reel on every platform and uses them to decide which formats to keep making. A format that works on Shorts and not on TikTok, or the reverse, is useful information, and it only exists if you post to both and measure both.

Four formats a day, two platforms, one pipeline

The biggest single gain in the experiment did not come from any one platform. When the account moved from one reel a day to four reels a day in four different formats, the daily TikTok total rose from about 800 views to about 1,700, with views per reel falling less than proportionally. Each format got its own test pool.

Now add Shorts. Four reels a day on two platforms is eight uploads a day, each with its own pool. For a human that is a full-time job. For a pipeline that writes, voices, edits and schedules, it is a configuration setting. The point of YouTube Shorts automation is not to save ten minutes per upload. It is to make the eight-pools-a-day strategy possible at all.

The volume that makes hits likely

The last piece is the one that decides whether any of this pays. Based on our analysis of 4.5 billion TikTok posts (2014 to 2026), among 2.87 million creators with at least 3 posts, the median number of posts before the first top 1% hit is 5, the third quartile is 16 and the 90th percentile is 38. The chance an account has at least one hit is 1.6% at 3 to 4 posts, 4.9% at 10 to 19, 10.9% at 20 to 49 and 39% at 100 or more. One percent of creators make 77% of hits.

Posting 1 to 3 times a week and 3 to 7 times a week give the same share of creators with a hit; above 7 a week the chance per post drops. So the target per platform is 3 to 7 a week, sustained for months. Shorts doubles the number of independent tests without doubling the creative work, which is as close to a free lunch as this data offers.

What to do with this

If your reels exist and are not on Shorts, that is the first fix, and it costs nothing. If they are 12 seconds long with a song underneath, that is the second fix, and it costs a script and a voice. If you post six and stop, nothing else matters.

Ultim handles the pipeline from URL to scheduled publish on TikTok, Instagram Reels and YouTube Shorts, with approval per reel or auto-publish, and 24 and 48 hour measurement on each platform. The plans start at 12 reels a month; details are at /pricing. The sensible first step is smaller: paste your website into /tools/reel-hooks and see whether the five hooks it returns are ones you would put a voice behind.