Interview clips make bad Shorts: how to make YouTube Shorts from podcast interviews
Interview clips make bad Shorts, because the best answer depends on a question the viewer never heard.
To make YouTube Shorts from podcast interviews that work, stop hunting through the full episode for a clip. Build the Short into the recording. Add a quick-questions segment at the end of every interview where the host sets up the context out loud and the guest gives a short, punchy answer. That clip stands on its own. And keep raw full-length episodes off your main channel, because their watch time drags down everything else you publish.
The raw-episode problem
On June 10 Amir, Charles and I went through why our clips weren't performing. We'd been doing what most B2B shows do: record the interview, upload the whole thing, then cut clips from it.
The first problem was the upload. A raw 45 or 50 minute interview on the main channel pulls the channel's overall watch time down. Most viewers don't sit through 45 minutes of a conversation between two people they don't know yet, so they leave early, and the channel's numbers show it. That low average then hurts everything else you publish next to it, Shorts included.
The fix there is simple. The full episode still has an audience: the guest, their network, the people searching for that exact conversation. It just shouldn't be the thing defining your main channel's numbers. Put it where long-form listeners already are, and keep the main channel for content built to be watched.
Why interview clips confuse cold viewers
The second problem was the clips themselves. Good Shorts are hard to pull out of an interview, even when the interview is great.
When we looked at why, it came down to one thing: context. The answer was great in context. In the interview, it came after twenty minutes of setup: who the guest is, what they built, the question that led there, maybe a story from earlier. A cold viewer scrolling Shorts gets none of that. They see a stranger mid-sentence, answering a question they didn't hear, referencing something they missed. They swipe.
Interviews are built for continuity. Shorts are built for strangers. The best moments of a good interview are the ones most dependent on everything around them, which makes them the worst raw material for a Short.
The quick-questions segment: designed for Shorts
So the fix we landed on changes the format instead of the editing. At the end of every episode, a dedicated quick-questions segment. The host gives the context up front, out loud, in one or two sentences: who the guest is and what the question is about. Then asks a sharp question. The guest gives a short, punchy answer.
Now every answer in that segment is a complete Short. It has a setup, a question and a payoff in under a minute, and the viewer needs nothing from the rest of the episode. Nobody has to dig through 45 minutes for a moment that might work. You record five or six moments that will.
It also helps the interview itself. The main conversation can go deep and wander where it needs to, without the host trying to manufacture clip-able lines the whole time. The Shorts have their own home at the end.
Some practical notes for running it:
- Say the guest's name and role in the setup, every time. The viewer doesn't know them.
- Ask questions with a clear tension: a myth, a mistake, a prediction, a number.
- Tell the guest the format before you start, so they answer in a few sentences instead of a few minutes.
- Frame vertically if you can, or leave room to crop.
Titles and thumbnails do most of the work
A week before that, on a call on June 2, the other half of this came up. Title and thumbnail optimization was called critical, about 80% of the success of a video, with A/B testing on thumbnails to find what converts. I'd say that's roughly right for long-form, and it's why the format change matters. A great title can get someone to click. Only a self-contained clip keeps them watching once they do.
For Shorts, the first second of the clip is your thumbnail. If the host's setup line is the first thing they hear, the viewer knows in one second what they're about to learn. That's the job a thumbnail does on long-form, done with your voice instead of an image.
Why a B2B show should care about Shorts at all
For us, the show is a pipeline before it's a media channel. Every guest is a relationship, and many are buyers or investors. So why bother with Shorts?
Because Shorts are what the next guest sees. When you invite someone onto a show, it's natural for them to look you up before they say yes. A channel full of 45-minute uploads with tiny view counts says nobody watches. A channel with sharp, self-contained clips of smart guests saying smart things says this is worth thirty minutes of your time. And every guest gets clips they can post, which they want, because they make the guest look good.
The quick-questions segment costs a few minutes per recording. It turns every interview into a set of assets that work for strangers, not just for people who already know you.
If you want the whole system, the show, the clips and the guests who turn into pipeline, take the tour of the show system.
Frequently asked questions
Should I upload full podcast episodes to YouTube?
Not as raw uploads on your main channel. Long episodes with low average watch time drag the channel's overall watch time down, which hurts everything else you publish there.
How do I make Shorts from an interview?
Build them into the recording. End each episode with a quick-questions segment where the host sets the context out loud and the guest answers in a few sentences, so each clip stands alone.
How long should a podcast Short be?
Long enough for a setup, a question and a payoff, and usually under a minute. If the clip needs earlier context to make sense, it's too dependent on the episode.
Do thumbnails matter for a B2B podcast?
Yes. On a June call we put title and thumbnail at around 80% of a video's success and A/B test thumbnails. For Shorts, the first line of the clip does the same job.