Three years ago I could open my inbox on a Monday and find four new shows asking about rates. Two of them would launch. One would still be publishing by episode ten. That was the boom, and everyone kept telling me it was about to end.
It did not end. It changed shape, and the people who called it a bubble were counting the wrong thing.
What actually collapsed was the number of new shows launched by people who had no reason to make one. The pandemic pushed a lot of curious hobbyists into buying a microphone, and most of them quit around episode seven, which is roughly where the novelty runs out and the work starts. If your measure of health is "how many podcasts exist", that number sagged and the headlines wrote themselves. But I sit on the editing side of this business, and from where I sit the last two years have been busier, not quieter. The difference is that the shows arriving now are serious. They have a reason to exist, a budget line, and someone whose job depends on the show performing.
The bubble was in launches, not in listening
A launch is cheap. Anyone can record a first episode. Sustained publishing is expensive in the only currency that matters, which is attention and time, and that is why the launch count is a garbage metric for the health of the medium.
Here is what I watch instead, because it is what shows up in my calendar:
- How many shows are on episode 40 or higher when they first contact me
- How many of them are attached to a business rather than a hobby
- How many are shooting on more than one camera
- How many ask about clips before they ask about the main edit
All four of those numbers went up. The market did not shrink, it got professional, and professional markets are harder to enter and much better to work in. A hobbyist compares my rate to free software. A company with a show compares my rate to the hourly cost of the person who was staying late to cut it themselves, and that comparison ends quickly.
Podcasts stopped being audio and nobody sent a memo
The thing that confuses the "boom is over" crowd is that they are still measuring an audio format. Most of my inbound now is video first. People record two to four cameras in a room, or a clean remote setup, and the audio feed is the byproduct rather than the point.
That single change rewired the whole economics of the format. Audio editing is a solved problem that a competent host can do alone on a laptop. Video is not. Video needs camera choices, eyeline logic, reframing, a consistent look across a season, and someone who can sit through ninety minutes of two people talking and know which of the four angles carries the moment. That is why our multi-cam editing work has grown faster than anything else we offer, and why the shows that stayed audio only are the ones feeling the squeeze.
If your show is one camera and one mic, that is still a completely valid format and we cut plenty of single camera shows. But you are now competing for attention against shows that look like television, and pretending that does not matter is how a good show stays small.
The economics quietly got better, not worse
There is a version of this argument where oversupply crushed everyone's rates. It did not happen in my corner of it. Prices held because the work got harder and the buyers got more sophisticated.
Our Starter tier is $199 an episode for single camera, 1080p, one revision. Studio is $299 for two to four cameras, 4K, and unlimited revisions. Both cover up to 60 minutes as a base, with 30 minute shows costing less and 90 minute shows costing more. A 4-episode pack saves 15 percent, monthly saves 25 percent, and most serious shows land on monthly within two months because publishing weekly and negotiating per episode is a miserable way to live.
What changed is the attach rate on everything around the edit. Vertical clips at $25 each, show notes and chapters at $35, rush 24-hour delivery at $75. Three years ago those were curiosities. Now the clips line is frequently the reason someone calls in the first place, and I would argue that clip work is where the highest leverage sits for a small show. The shape of it is simple: the edit is the floor, the distribution assets are what make the edit pay for itself.
The shows that quit were not beaten by the market. They were beaten by the fact that publishing weekly is a job, and nobody warned them.
Why I think the "AI killed editing" story is backwards
Every few months someone forwards me a tool that promises to cut a podcast automatically. I have tried most of them, and I wrote down what a year of that experimenting actually taught me in a longer piece on AI editing tools. The short version is that these tools are genuinely good at the mechanical layer and genuinely bad at the layer that makes an episode watchable.
Silence detection can find a gap. It cannot tell that the gap is a guest deciding whether to say the honest thing, which is the best three seconds in the episode. That is why nothing we deliver is auto-cut by silence detection. A human editor watches the whole episode, start to finish, every time. It is slower and it is the entire product.
The irony is that AI tooling made the boom bigger, not smaller. Cheap transcription and cheap cameras lowered the cost of producing a show, more shows exist at a serious level, and more serious shows means more demand for the judgment that software still cannot fake. Most of that judgment is invisible, which is why I keep arguing that the editing decisions nobody notices are the ones doing the heavy lifting.
What a mature market actually feels like from the inside
I have cut past a thousand episodes at this point, and the pattern that separates shows that last from shows that fade has almost nothing to do with the market and almost everything to do with the operating rhythm. I collected the honest version of that in what a thousand episodes taught me, but the short list looks like this.
Shows that survive tend to:
- Record in batches, so a bad week does not become a missed week
- Keep a fixed publish day and treat it like a deadline, not a target
- Decide the format once and stop redesigning it every quarter
- Ship clips from every episode instead of only the good ones
- Fix their room before they buy another camera
Shows that fade tend to redesign the intro four times, chase a guest who will not commit, and go quiet for a month because one episode was hard. None of that is a market problem. It is a habits problem wearing a market problem costume.
Our first cut lands in 48 hours and revisions come back the next working day, and that is deliberate. A fast turnaround is not a luxury feature. It is the thing that lets a host keep a rhythm when the week goes sideways, and rhythm is the whole game.
Where the growth actually is right now
If I had to bet money on where podcast demand goes over the next two years, I would put almost all of it on one thing: shows built for a business rather than for an audience number.
A show with 800 listeners that produces four qualified conversations a month is a better business than a show with 40,000 listeners and no offer. That is the shape of most of the work coming in now, and it is why LinkedIn podcast clips turned into one of the most requested things we do. The clip is the show for most of the audience. The full episode is the proof for the small slice that goes looking.
The second place I would bet is packaging. A video podcast lives or dies on the thumbnail long before anyone judges the edit, which is why our sister studio ClickTheory exists and does 8 thumbnails for $250. I have watched identical episodes perform three times differently based on packaging alone. That is not an editing problem, and it is worth knowing which problem you actually have.
The part where I say the unpopular thing
I turn down work fairly often, and the reason is usually that the show is not ready to be edited by anyone. I wrote about that in why I turn down some podcast clients, and it is the same argument as this whole post: the constraint is almost never the market.
If you are three episodes in and wondering whether it is too late, it is not too late. It was never a race with a start gun. What is true is that the standard went up, and a show that would have passed in 2021 now reads as unfinished. The good news is that the bar is mostly about consistency and craft, both of which are buyable and learnable. If you want to cut it yourself, the walkthrough on how to edit a podcast in Premiere Pro is genuinely the process I use, not a watered down version of it.
The boom is not over. The tourists went home. What is left is a room full of people who intend to still be publishing in three years, and honestly, that is a much better room to be in.
If you are in that room and the editing is the part slowing you down, write to me at samy@pixelsamy.com and tell me what your week actually looks like. I will tell you straight whether we are the right fit.