PodPolish

Record it right. Fix less. Know the bill.

Podcast Editing Software: A Cost-Per-Episode Worksheet

We costed a 60-minute show published four times a month. The biggest line was never the app — it was the noise work.

from the PodPolish benchpublished September 4, 2026

Pick podcast editing software on cost per episode, not on the feature grid. We run a 60-minute conversation show four times a month, and when we costed our own workflow the surprise was not the software line. Audacity is free and open source and will do the whole job if you are patient with it. Descript's free tier gives 60 minutes a month. Adobe Podcast Enhance gives an hour a day in the browser. Below is the worksheet we actually use. Fair warning about where it lands: the biggest line item was never the app. It was the two hours a week we spent scrubbing a co-host's air conditioning out of a take that was otherwise fine.

Cost per episode, not cost per month

A monthly price tells you nothing until you divide it by episodes. Four shows a month at 60 minutes each is roughly four hours of raw audio, and more if you record video or keep a separate track per guest. Run that number against a free tier and the ceiling shows up immediately: a monthly allowance of 60 minutes covers about a quarter of one episode, which is a taste rather than a workflow.

Two things fall out of that list. First, only one option has no ceiling at all, and it is the one with the steepest first week. Second, the metered tools are metered on what you send them, not on what you publish, so a 90-minute raw file costs more allowance than the 60-minute episode it becomes. Trim before you upload. It sounds obvious written down and we still forget it about once a quarter, usually at eleven at night.

Where the hours actually go

We timed ourselves for two months. Cutting and structure took less time than expected. Noise work took much more: a hum on one channel, an air conditioner cycling on and off, a laptop fan that only shows up in the quiet parts between answers. None of that is a hard edit. It is just slow, and it scales with how bad the recording was rather than how long the episode is. A clean 60-minute take edits faster than a noisy 40-minute one, every time, without exception in our logs.

Repair has a ceiling too

There is a point past which no processing helps. Push a denoiser hard and the consonants go first — the s and t sounds thin out, then the whole voice starts to sound like it is arriving through a wall. On our bench the rule is simple: if a fix is audible as a fix, we go back to the source instead. Which is a comfortable thing to say right up until the source is a guest in a spare room with a window open onto a main road, three time zones away, and the recording is already done.

The worksheet, line by line

Start with the raw hours. Four episodes at 60 minutes is four hours of audio in, and if you record a separate track per person, double or triple that. Now assign each hour to a tool. Our own numbers live on the cost per episode page and they move as tiers do, but the shape holds: free-forever tools cost time, metered tools cost planning, paid tools cost money and save both.

Then the routing. We record locally for each person rather than taking the call recording, because a local track survives a bad connection and a call recording does not. The full method is in our recording guide. Cleanup goes one of three ways depending on the damage: the Adobe Enhance workflow for a browser pass on a rough file, the Descript Studio Sound workflow when the edit is already happening there, and the Cleanvoice workflow when the problem is filler words and crosstalk rather than room tone.

Last line, the one that pays for itself: whatever you can stop recording in the first place. Mic technique, a rug, a closed window, and live suppression on the noisy end of a remote interview. Our guide to removing background noise from a podcast covers what to do when prevention did not happen, which is most weeks. Add the whole worksheet up per episode and the ranking usually flips away from whichever app has the longest feature list.

Where the worksheet breaks

Video changes every number

A video podcast multiplies file sizes and pushes you toward tools that handle both tracks together. Upload ceilings measured in megabytes bite first — 500 MB is not much video — so the browser route stops being practical and the local editor gets more attractive. Budget the storage as well as the software.

One noisy guest is not a software problem

If nine episodes are clean and the tenth is a mess, do not re-plan your whole stack around it. Fix that episode with whatever gets it over the line, then go back to the normal routine. Our podcast audio editing notes cover the ordinary case, and the exception is allowed to be expensive once in a while.

Free tiers change without warning

Every number here was checked in August 2026 and every one of them is a moving target. Before you build a monthly routine on a free allowance, open the pricing page yourself and confirm the ceiling. The tools that have not moved in years are the open-source ones, which is an underrated reason to keep Audacity installed even when it is not your main editor.

Questions we get

What is best for beginners?

Whatever you will actually finish an episode in. Audacity is free, open source and does everything a talk show needs, at the cost of a fiddly first week. A cloud editor with a free tier is gentler but metered, so check the monthly allowance against your episode length before you commit a whole season to it.

Is there genuinely free podcast editing software?

Yes. Audacity is free and open source with no ceiling and no account. Everything else described as free is a metered tier: Descript gives 60 minutes a month, and Adobe Podcast Enhance gives an hour a day in the browser with files capped at 30 minutes or 500 MB. Those are trials of a workflow rather than the workflow.

Can AI tools clean up a bad recording?

Partly, and with a limit. Model-based cleanup handles steady room noise well and struggles with a second voice, with music, or with a recording that was clipping at the source. Push it hard and consonants thin out. Treat it as a rescue for a take you cannot record again, not as permission to record badly.

Which do we use for a weekly show?

A local editor for the cut and a separate pass for cleanup, because the two jobs want different tools. The specific apps matter less than the order: fix the noise first, then cut, so you are not making a hundred small decisions against a background that keeps shifting under you.

Does better software beat a better recording?

No, and it is not close. An hour saved by a quiet room beats an hour of processing every time, because repair always costs some quality and prevention costs none. Spend the first budget on the room and the mic position, the second on live cleanup for remote guests, and only then on the editor.