How shorts cut from your own videos feed the channel that made them

If you publish long videos, clipping them is the cheapest content you will ever make. It is already filmed, already yours, and there is no rights question to think about.

What surprises people is that a hundred thousand short-form views can produce almost no new watch time on the long videos. That is not a failure of shorts. It is a failure of what went into them.

A clip is a trailer, not a summary

The clip that performs best in isolation and the clip that drives people to the source are often not the same clip. A complete, self-contained answer gets saved and shared, and gives the viewer no reason to go anywhere.

The version that works keeps something back. It establishes that something interesting is being discussed, gets to the edge of the answer, and stops. Not a cliffhanger with nothing behind it, which viewers resent, but a genuine "there is more to this" that happens to be true.

The test: after watching your clip, does the viewer have a question? If they have only an answer, the clip was the destination rather than the door.

What never to put in a clip

• The single best moment of the video, in full. Save it. It is the reason to watch the source.
• The conclusion of anything you spent the video building towards.
• Anything that only makes sense if you saw the previous twenty minutes. It will not travel.

What to use instead is the setup, the surprising claim, the disagreement, the moment somebody says "that is not what I expected". Those work standalone and they point forwards.

Every long video should produce more than one kind of clip

Cutting six clips of the same shape from one video wastes the source. A long video usually contains three different jobs:

• The hook clip. The most arresting thirty seconds, aimed at people who have never heard of you. This one is for reach.
• The useful clip. A self-contained tip that stands entirely on its own. This one is for saves and follows, and it is fine that it gives everything away.
• The trailer clip. The one that deliberately stops short and points at the full video. This one is for watch time.

Most creators only ever cut the first kind, then conclude that shorts do not convert.

Short-form viewers are not long-form viewers yet

Someone who watched thirty seconds of you in a feed has not decided anything. Expecting them to commit to a forty minute video immediately is asking for a jump most people will not make.

What moves them along is repetition. Seeing three clips from the same person over a couple of weeks does more than any single call to action, which is an argument for clipping consistently rather than in bursts whenever you remember.

How to tell whether it is working

Short-form views are the wrong number to watch. They tell you the clip travelled, not that it did its job. The numbers that answer the actual question are:

• Subscribers gained from short-form, which is the funnel working slowly.
• Views on long videos from browse and suggested after a clip run, which is the funnel working properly.
• Whether the specific videos you clipped got more traffic than the ones you did not.

That last one is the closest thing to a controlled test you will get. Clip half your uploads for a month and compare.

The workflow

• Track your own channel so every upload gets clipped the day it goes out, while the video is still being pushed.
• Cut more clips per video than you plan to post, then choose between them.
• Make sure at least one is a trailer clip rather than a complete answer.
• Space them across the following days instead of posting the batch at once.
• Check long-form traffic monthly, not short-form views weekly.

Clipping your own work is the rare case where the volume is free and the only real decision left is what you choose to give away.

Common questions