A host messaged me last month with a screenshot. Six clips posted in three weeks, 180,000 views across them, and eleven new subscribers on the show. Eleven. He wanted to know what was wrong with the clips.
Nothing was wrong with the clips. That was the problem.
I have cut a lot of vertical clips at this point, and I have watched the same pattern show up on shows in fitness, in B2B software, in property, in therapy. Views go up. Subscribers do not move. And the host concludes that short form is a scam, or that the algorithm hates him, or that he needs a better editor. Usually it is none of those. Usually the clip did its job perfectly and the job was the wrong one.
Views are an attention metric, not an interest metric
Here is the uncomfortable part. A view on a vertical clip means a thumb slowed down for a second and a half. That is it. It is closer to a billboard impression than to a magazine subscription.
A clip that goes viral proves your hook works. It proves nothing about whether anyone wants forty more minutes of you. Those are two separate purchases, and most hosts only ever make the first one.
When I look at a clip that pulled big numbers and zero follows, I am usually looking at a piece of content that was complete in itself. The hot take landed. The story resolved. The viewer got the whole thing in 38 seconds and felt satisfied. Satisfied people do not subscribe. Curious people subscribe.
The clip has to leave something on the table
This is the single change that moves the number, and it costs nothing.
Stop cutting to the most complete 45 seconds. Start cutting to the most incomplete one that still makes sense on its own.
There is a real difference. A complete clip answers the question. An incomplete clip answers half of it, well enough to be satisfying as a piece of media, while making the other half feel expensive to lose. The guest says the thing that took him nine years to learn, and you cut right before he explains how he applies it on a Monday morning.
I am not talking about cliffhanger bait. Nobody follows a show because the clip said "and the answer will shock you". I mean genuine unfinished business. The clip should end on a moment where a normal person thinks: wait, what did he do next.
The clip is not a sample of the episode. It is an argument that the episode exists and is worth an hour of a stranger's life.
That reframe changed how I mark clips inside a cut. I no longer look for the funniest moment. I look for the moment that is impossible to fully understand without the surrounding twenty minutes, then I cut just enough of it to be watchable alone.
Nobody knows what your show is called
Second failure mode, and it is even dumber than the first.
I have watched clips with a quarter million views where the show name appears nowhere. Not in the caption burn-in, not in the frame, not in the first line of the description, not in the audio. The viewer laughed, scrolled, and had no idea they had just watched a podcast at all. They thought it was a guy in a chair.
Fix list, in order of how much they matter:
- A small persistent show mark in a corner of the frame, not a giant logo bug. It needs to be legible on a 6 inch screen at arm's length, which usually means bigger than you think and quieter than your designer wants.
- The show name spoken out loud somewhere in maybe one clip in four. Audio identity beats visual identity on platforms where people watch with sound on.
- The same caption style on every single clip. Same font, same size, same position, same colour. This is the cheapest brand asset you will ever own and most shows change it every month.
- A profile that says what the show is in five words, so the one person in a hundred who taps through actually converts.
That last one is where most of the leak sits. People do tap through. They land on a page with a blurry logo and a bio that says "conversations with interesting people", and they leave. You paid for that traffic in editing hours and you lost it in a bio.
Consistency is the actual subscribe trigger
I have come round to a fairly blunt view here. People do not subscribe to clips. They subscribe to patterns. One great clip gets a view. The third clip they recognise gets the follow.
That means the frequency question matters more than the quality ceiling. Four decent clips a week from one episode will out-recruit one perfect clip a month, every time, because recognition needs repetition. This is also why I push hosts towards turning one episode into a month of assets rather than treating every clip as a bespoke project. A month of consistent presence beats a spike.
We price vertical clips at $25 each as an add-on for exactly this reason. It is not meant to be a premium boutique item. It is meant to be cheap enough that you order six from every episode without thinking about it, because six is where the pattern starts to work.
Where the subscribe actually happens
Almost never inside the clip. Usually one step later, and the step is different on every platform.
On short form feeds, the follow happens on the profile page after a tap. So the profile has to be built for that visit: a pinned clip that explains the show, a clean grid, an obvious link.
On YouTube, the subscribe happens on the long form video. A Short can send someone to your channel page, and your channel page has to have a watchable episode with packaging that survives a two second glance. This is the part hosts skip. They spend four hours on clips and use a screenshot of the Riverside recording as the thumbnail. If your clips are working and your long form is not converting the traffic, the thumbnail is usually the bottleneck, which is why our sister studio ClickTheory exists and does eight thumbnails for $250.
On LinkedIn it is different again, because the follow tends to come from the person, not the show, and the comments do more work than the video. I wrote more about how that platform behaves in our notes on LinkedIn podcast clips, but the short version is that native uploads with burned captions and a real opinion in the post text outperform anything that looks syndicated.
The quality floor that people actually notice
I want to be careful here, because "improve your quality" is lazy advice. Most shows do not have a quality problem, they have a packaging problem.
But there is a floor, and below it nothing else you do matters. Audio that clips. Faces lit from one side with a hard shadow across the eye. A frame where the guest is half out of shot. Those things read as amateur in under a second, and a stranger deciding whether to follow you makes that call fast. Lighting in particular gets blamed on the camera when it is almost always a placement issue, and we have a whole piece on the lighting mistakes that show up in the edit because we keep receiving the same three problems.
Resolution matters less than people think for clips, because vertical delivery compresses hard anyway. It matters more for long form and for reframing, which is the honest version of the 4K question. If you shoot 4K you can punch in and reframe a clip without losing anything. That is the real argument, not sharpness.
What I would do with a show that has views and no subs
If someone handed me a channel in this exact state, here is the order I would work in.
- Fix the profile and the packaging first. It is one afternoon and it changes the conversion rate on traffic you are already getting.
- Rebuild the clip selection rule. Incomplete over complete. Curiosity over closure.
- Lock one caption and frame style and do not touch it for ninety days.
- Raise clip volume before raising clip polish. Six a week beats two.
- Only then look at the long form edit, the multi-cam editing rhythm, the intro, the pacing.
Most hosts do that list backwards. They start with the intro animation.
The part nobody says out loud
Some episodes do not deserve clips. If the conversation was flat, no editor can manufacture a hook from it, and the ones we cut anyway are the ones that get 900 views and nothing else. A human watching the whole episode can tell you that. Silence detection and auto-clip tools cannot, which is why we do not use them: every episode we cut is watched end to end by an editor, and part of what you are paying for is someone saying "there are three real moments in this one, not eight".
That honesty is worth more than the extra five clips.
If you want a second pair of eyes on why your clips are pulling views and no follows, send me an episode and I will tell you what I would cut and why. I am at samy@pixelsamy.com, based in Dubai, working with shows all over the place. Or read how it works first if you would rather see the process before you talk to anyone.