Turn Long Videos Into Shorts

The single biggest failure in repurposing is clipping something that made sense in context and doesn't make sense out of it. A brilliant point that opens with 'and that's exactly why it happened' is unusable, because the viewer has no idea what 'it' is. Self-containment is the first filter, and this tool penalises segments that start mid-thought for exactly that reason.

After that, the traits that predict a good clip are fairly consistent. A concrete number gives the clip a spine and reads as evidence. A question functions as a built-in hook. A claim that contradicts what the viewer assumes generates comments. A short story holds attention because people finish narratives they've started.

Length matters more than people expect. Somewhere around 30 to 60 seconds of speech is the sweet spot: long enough to say something complete, short enough that retention stays high. A 90-second clip has to be exceptional to hold, because retention is measured proportionally and the bar rises with duration.

If you already produce long-form content, you're sitting on the most expensive input already paid for. The research is done, the argument is made, and the delivery is recorded. Cutting it into shorts is close to free content relative to originating something new.

It also solves the consistency problem. The reason most short-form channels stall is that daily production is a job, and repurposing turns one recording session into a week or more of posts. A single hour-long episode can reasonably yield five to eight clips.

The distribution logic works too. Short-form is discovery, long-form is depth. A clip that performs sends viewers back to the full video, which is where the higher ad rates and the actual relationship live, particularly on YouTube where a Shorts viewer can become a long-form subscriber.

A 16:9 recording cropped naively to 9:16 loses most of the frame, and if two people are talking it loses one of them. The usual fixes are speaker-tracked cropping that follows whoever's talking, or a stacked layout with the video on top and captions or b-roll filling the lower third.

Captions are not optional. A large share of short-form viewing happens muted, and a clip from a podcast is almost entirely speech, so without captions there's nothing for a muted viewer to engage with at all. Burned-in captions, present from the first frame, not a caption track.

Keep the safe zones in mind when you compose. All three platforms overlay their own UI across the bottom of the frame, and TikTok also uses the right edge. Anything important down there gets covered.

Record or publish the long video first, then pull the transcript and run it through this tool. Take the top five candidates rather than all ten, since the tail is usually weaker and posting mediocre clips trains your audience to scroll past you.

Rewrite the opening of each clip. The transcript gives you the substance, but the first line of a short needs to be a hook, and a sentence that worked as the tenth minute of a podcast rarely works as the first second of a Short. This is where most of the value is added.

Then produce each one in the right shape for each platform rather than exporting once and uploading three times. Different openings, no third-party watermarks, captions burned in. That's the part worth automating, because it's mechanical and it's the part people skip when they're busy.