A one-hour podcast holds about 29 moments that work as a short on their own. That is the median across 187 long videos people ran through Skapo between August and November 2026. Nobody posts all of them. But it tells you whether one episode a week can carry a clip a day (it can, with room to spare) and how long those clips tend to be (about 43 seconds).
Skapo clip yield benchmarks, 2026
- Usable clip moments per hour of source video: 28.6 (median, 187 videos)
- 45-minute episode: 20 moments (middle 80%: 14 to 30)
- 70-minute episode: 36 moments (middle 80%: 25 to 44)
- Median finished clip length: 43 seconds (10th to 90th percentile: 32 to 58 s)
- Share of clips between 30 and 60 seconds: 89% (3,855 clips)
Source: Skapo (skapo.io) production data, 119 accounts, August 9 to November 9, 2026. Cite as “Skapo 2026 clip yield benchmarks”.
How many clips per episode, by episode length
A “usable moment” here is a stretch of the conversation that opens with a hook, makes sense to someone who never heard the episode, and ends on a finished thought. That is the bar a clip has to clear before Skapo will offer it. The count below is how many cleared it, before any plan limit on how many get rendered.
| Episode length | Typical | Videos | Usable moments | Middle 80% |
|---|---|---|---|---|
| 10 to 30 min | 21 min | 78 | 10 | 5 to 16 |
| 30 to 60 min | 45 min | 58 | 20 | 14 to 30 |
| 60 to 90 min | 70 min | 29 | 36 | 25 to 44 |
| 90 to 150 min | 102 min | 22 | 48 | 29 to 50* |
* The shortlist stops at about 50 moments per video, so for episodes over 90 minutes the top of the range is a floor, not a ceiling.
Two things stand out. First, the yield is steady at roughly one usable moment every two minutes once an episode passes half an hour, so doubling the length about doubles the clips. Second, short videos give less than that rate suggests. A 21-minute video gave 10 moments, not 14, because the intro, the sponsor read and the sign-off take a bigger share of a short video and none of them make a clip.
How long podcast clips come out
Across 3,855 clips rendered in 2026, half run 30 to 45 seconds and another 39% run 45 to 60. Very few are under 20 seconds. That is too little time to set up a point and land it.
| Clip length | Share of clips |
|---|---|
| Under 20 s | 0.8% |
| 20 to 30 s | 3.2% |
| 30 to 45 s | 50.1% |
| 45 to 60 s | 38.9% |
| 60 to 90 s | 6.1% |
| 90 s and up | 0.8% |
These lengths come from where each story starts and ends, not from performance data. So read them as “how long a complete thought usually takes in conversation”. They say nothing about how long viewers watch. The platform rules worth knowing:
- YouTube Shorts accepts vertical videos up to 3 minutes.
- Instagram Reels accepts up to 3 minutes.
- TikTok takes much longer uploads, but its Creator Rewards Program only pays on videos over 1 minute. If TikTok payouts matter to you, pick the moments that run past 60 seconds.
How many clips to actually post
Usable is not the same as worth posting. Of 20 to 40 usable moments from an hour, a few will be clearly the best, some will be good, and the rest will cover the same point twice. Here is a rhythm that works for a weekly show:
- One clip a day, seven a week, from the latest episode. Lead with the strongest on the day the episode drops.
- Keep the next five or so as backup for weeks where an episode is thin or a guest cancels.
- Post the same clip to Shorts, Reels and TikTok. One cut, three places. Captions and framing that work on one work on all three.
If you clip someone else’s show through a clipping campaign, the math changes. You are paid on views, so volume across many episodes beats depth in one. Take the top three or four from each episode and move on to the next upload. Check what the campaign allows before you post. See whether clipping other people’s content is legal and what clipping campaigns pay.
What gives you more (or fewer) clips
- Opinions and stories beat logistics. A guest who tells a story with an ending gives you a clip. Twenty minutes on scheduling and plugs does not.
- Questions that get a full answer. Interview formats where the host asks one clear question at a time give far cleaner clips than crosstalk.
- Every speaker on camera. A wide shot with everyone visible lets a clip keep the reaction as well as the answer. See the split-screen layouts for 2 to 4 speakers.
- Audio-only shows give zero. Shorts need a video recording. Without one, the closest thing is an audiogram, which is a different format.
How these numbers were measured
Moment counts cover every video of 10 minutes or longer that a Skapo user ran as a full job between August 9 and November 9, 2026: 187 videos from 119 accounts. They include podcasts, interviews, streams and talk-show clips, both people’s own shows and other creators’ videos. Partial runs that analysed only a slice of a video are left out. Medians are shown with the 10th and 90th percentiles, because an average hides how much episodes vary. Clip lengths cover all 3,855 clips rendered from April 21 to November 9, 2026. Render times and completion rates from the same production data are on the benchmarks page.
Find out how many your episode has
Paste a YouTube link or upload an episode of up to two hours. The free plan returns the top 7 clips with captions and speaker framing in about 10 minutes, no card needed.