Last month I pulled the numbers on a show we cut every week. Twelve full episodes on YouTube. Roughly forty vertical clips across TikTok, Reels and Shorts. The episodes did about nine thousand views combined. The clips did just over four hundred thousand.
The host looked at that and said the clips were "working" and the podcast was not. I think he had it exactly backwards.
The clips were not the win. The clips were the door. Nobody wakes up wanting to watch a ninety minute conversation with a stranger. They meet you in eight seconds on a phone, decide you are not annoying, and then maybe, weeks later, they give you the long version. Your clips are not marketing for your podcast. Your clips are the top of your funnel, and the episode is the middle.
Once you accept that, a lot of decisions get easier and a lot of arguments go away.
The episode is not the product, the trust is
Hosts tell us they want more downloads. Fair. But downloads are a lagging measure of something slower: how many people currently believe you are worth an hour of their life.
That belief does not get built in an hour. It gets built in dozens of small, cheap encounters. A clip where you say something slightly risky. A clip where you disagree with a guest and do it well. A clip where you explain a thing nobody else explains properly. Each one is a tiny deposit.
The full episode is where the deposit gets withdrawn. It is the thing that converts a person who recognises your face into a person who knows your point of view. That is why I hate the phrase "repurposing". Repurposing implies the clip is a leftover. It is not a leftover. It is the first thing almost everyone will ever see of you.
If the first thing is an afterthought, so is your growth.
Treat the clip as the ad, not the appetiser
Here is the mental shift I push on every host I work with. Stop asking "what is a good moment from this episode". Start asking "what is a good ad for this show".
Those are different questions and they produce different clips.
A good moment is subjective and usually sentimental. The host always loves the bit where the guest got emotional, because he was in the room. A good ad has a job: stop a thumb, land one idea, and leave a specific impression of who you are.
Which means most clips should be built from one of four things:
- A claim someone will disagree with, stated plainly in the first line
- A number that sounds wrong until it is explained
- A specific story with a turn in it, not a general reflection
- A short piece of practical instruction that works on its own with no context
Notice what is missing. Introductions. Guest credentials. "So tell us about your journey." Nobody has ever subscribed to a podcast because a clip opened with a job title.
A clip that needs the episode to make sense is not a clip. It is a fragment, and fragments do not travel.
This is also the honest reason a lot of clips get views and nothing else. If the clip is entertaining but says nothing about what your show is, you rented attention for eight seconds and gave it back. I wrote more about that failure mode in why your clips get views but no subscribers, because it is the single most common complaint that lands in my inbox.
Funnels need a middle, and yours is probably broken
Top of funnel is the clip. Middle is the episode. Bottom is whatever you actually want: a newsletter, a product, a sponsor deal, a client.
Most shows I see have a fat top and a broken middle. Great clips, then the viewer taps through to the channel and finds a full episode with no cold open, a two minute intro animation, and a thumbnail that looks like it was made in a hurry. The handoff fails right there.
So the middle deserves as much attention as the clip did:
- The first thirty seconds of the episode should be the best thirty seconds, not the politest
- The thumbnail and title should promise the same thing the clip promised
- Chapters so someone can find the part they came for
- Consistent branding so the clip and the episode feel like the same show
Packaging is doing more work than people admit. When a host asks me why a strong episode underperformed, the answer is usually the frame, not the cut. That is why we run ClickTheory as a sister studio for thumbnails. Eight thumbnails for $250 is a small amount of money next to the cost of a good episode landing on nobody.
For the cut itself, this is exactly why we do not automate. A human editor watches the whole episode before a single clip gets pulled, because you cannot find the ad inside the conversation by looking at a waveform. Silence detection finds gaps. It does not find the sentence that will make a stranger stop scrolling. I get into the texture of that difference in why auto-cut podcasts feel wrong.
Volume matters, but only after the format is right
I am not going to pretend distribution is a purely creative problem. It is partly arithmetic.
If you post two clips a week, you get roughly a hundred shots a year to be discovered. If you post six, you get three hundred. The hit rate per clip is low and unstable for everyone, including people with big accounts, so the number of attempts is doing a lot of the work.
But there is an order of operations here that people get wrong. Volume amplifies whatever you already have. If your format is muddy, more clips just means more people forming a vague impression of you. Nail the format on five clips, then scale to twenty.
Practically, for most weekly shows, that looks like four to six verticals per episode. We price them at $25 each as an add-on, and my honest advice is to start at four and only go up when you can point to two clips that clearly worked and explain why.
The consistency problem nobody warns you about
Here is the part that has nothing to do with editing and everything to do with whether this works.
Funnels are cumulative. A clip you posted in March is still bringing people in during July. Which means the compounding only happens if you are still there in July. And most hosts are not. The drop-off is brutally consistent and it has a shape I have watched enough times to predict, which is the whole reason I wrote why most podcasts quit at episode 7.
The failure is almost never a lack of ideas. It is that the production load is front-loaded onto the person who also has to host, book guests, and run a business. Recording is the fun part. Cutting forty clips at eleven at night is not.
So build the pipeline you can actually sustain:
- Batch recording. Two or three episodes in a day, not one a week
- Fixed publish day, no exceptions, even for a weaker episode
- Editing off your own plate before you start resenting it
- One clip format that you can repeat without redesigning it every time
Our first cut comes back in 48 hours and revisions the next working day, which exists for exactly this reason. The bottleneck in most podcasts is not talent. It is turnaround.
What sponsors are actually buying
The other reason to take clips seriously is money, and this is where the funnel framing becomes very literal.
Sponsors do not buy downloads any more, or at least the good ones do not. They ask what the audience does, whether the host has a recognisable point of view, and whether the show shows up in places where people already are. A show with modest downloads and clips that consistently reach a specific audience is a far easier sell than a show with a bigger number and no visible surface.
I broke down that conversation in more detail in what sponsors look at before they pay, but the short version is that clips are the evidence. They are the only part of your show a sponsor can evaluate in two minutes.
Getting the raw material right
None of this works if the source footage fights you. A clip lives or dies on the face being sharp, the audio being clean, and the framing surviving a crop to vertical.
Two things I say constantly. First, record wider than you think you need, because a 4K wide shot gives an editor room to punch in for a vertical crop without turning your face into mush. Second, get the audio right at the source, since no amount of cutting rescues a room with an echo.
If you are choosing tools, the Riverside vs Zoom comparison covers the practical difference in what you end up with on the timeline. And if you are weighing who cuts it, the Fiverr editors versus an agency breakdown is the argument I would make in your position, not mine.
On our side, Starter is $199 an episode for a single camera at 1080p with one revision, and Studio is $299 for two to four cameras at 4K with unlimited revisions. Both cover up to 60 minutes, shorter is cheaper, longer costs more. Packs bring that down: four episodes saves 15%, monthly saves 25%. You can see how clips slot into that on the service page rather than me listing it all again here.
Where I would start on Monday
If you do one thing this week, do not add more clips. Pick your last ten and sort them honestly into two piles: ones that would make a stranger understand what your show is, and ones that would not.
Most people find the second pile is much larger. That is the actual problem, and it is fixable without spending anything.
Then fix the handoff. Make sure someone who arrives from a clip lands on an episode that opens strong and looks like it belongs to the same show. That single repair is worth more than doubling your posting schedule.
I am in Dubai and we work with hosts anywhere. If you want a second opinion on whether your clips are pulling their weight, send me an episode at samy@pixelsamy.com and I will tell you what I would cut and why. No pitch attached.