Zencastr vs Riverside: which one is actually easier to edit into a video podcast?

Pick Riverside if video is the point of your show, and pick Zencastr if audio is the point and video is a bonus. That is the short answer, and after cutting a lot of both, I have not found a reason to complicate it.

Riverside records locally on each participant's machine at up to 4K per person, then uploads those files in the background. You get separate video tracks and separate audio tracks per guest. That separation is the whole ballgame for editing. It means I can punch in on the guest during their answer, cut a bad take out of the host track without touching anyone else, and fix one person's room noise without making the other person sound processed.

Zencastr came from the audio world and it still shows in the best way. The audio pipeline is careful, the WAV tracks are clean, and the interface does not fight you when all you want is a talk show. Video works, and it is fine, but it feels like the second thing the product does rather than the first.

So the question is not which tool wins. It is what your show is. If you are publishing full episodes to YouTube with two people on screen for 45 minutes, Riverside gives me more to work with. If you are publishing to Spotify and Apple with a video version as a nice extra, Zencastr costs you nothing in quality where it counts. Either way the recording is only step one of podcast post production, and that is where most of the perceived quality actually gets made.

It arrives as a folder of per-participant files, and the difference between the two platforms is mostly about how much of that folder is usable.

From Riverside I usually get one MP4 or MOV per participant plus separate WAV audio, all starting at the same reference point. Frame rate is consistent. Local recording means the video is not a compressed stream of whatever the internet allowed that day, so when I zoom in 15% for a reaction shot it still holds up. If a guest's connection dropped mid call, the local file usually survived even though the live preview stuttered.

From Zencastr I get excellent separate audio and a video track that is perfectly serviceable at 1080p. For a talking head show cut in a simple two-up or speaker-switch layout, nobody watching can tell the difference. Where it gets tight is when you want to crop aggressively for vertical clips, because cropping a 1080p source to 9:16 leaves you very little pixel budget.

One thing both platforms get right, and I want to be fair here: separate tracks. Any recorder that gives you a single flattened mix has already cost you more in editing time than the subscription saved. Both of these pass that test.

The practical checklist before you hit record on either one: everyone on wired internet if possible, headphones on every participant so there is no bleed, a lamp in front of faces rather than a window behind them, and a hard clap at the top so I have a visual sync point if anything drifts. That last one takes two seconds and has saved me an hour more than once.

Both ship an editor now, and both are genuinely useful for a fast turnaround, with the same ceiling.

Riverside's editor does text based cutting, filler word removal, and automatic clip suggestions. Zencastr has moved the same direction with transcript editing and highlight generation. If your goal is to publish a raw conversation within an hour of recording, either one will get you there and the result will be acceptable.

Where they both plateau is judgment. An automated clip picker finds moments where the transcript looks quotable. It does not know that the guest's best line landed 20 seconds later, after a pause, or that the setup sentence before the punchline is what makes the punchline work. It also does not know your show. When we build podcast highlight reels, the person choosing the moment has watched the whole episode and knows which idea you actually want people to associate with you.

Same story with filler word removal. Cutting every single "um" mechanically leaves speech that sounds slightly wrong in a way viewers feel but cannot name. A human leaves the breaths that make you sound like a person and cuts the ones that make you sound unsure.

Audio cleanup follows the same pattern, and I wrote about that tradeoff in more detail in Auphonic vs human audio mixing. Automated leveling is genuinely good at consistency. It is not good at deciding that the guest's laugh should stay loud.

Both platforms sit in a similar subscription range, both charge more for higher resolution and more recording hours, and both are cheap relative to what happens next. The recorder is rarely the expensive part of a podcast.

The cost that surprises people is the editing. If you hire a video podcast agency, you are usually looking at $300 to $400 per episode for a standard cut, plus roughly $400 a month once you add a steady stream of vertical clips. That is on top of your recording subscription.

We charge $199 per episode on Starter, which covers a single camera show at 1080p with one revision round. Studio is $299 per episode and covers 2 to 4 cameras at 4K with unlimited revisions, which is the plan most Riverside shows land on because Riverside is handing us multiple angles worth using. Both are priced for episodes up to 60 minutes. Under 30 minutes costs less, up to 90 minutes costs more.

Buy in bulk and it drops further. A 4-episode pack saves 15%. Monthly saves 25%. Add-ons are flat and unbundled so you are never paying for something you did not ask for: vertical clips at $25 each, show notes and chapters at $35, and rush 24-hour delivery at $75 when a guest posts earlier than planned. Full numbers live on the pricing page.

One more thing that costs nothing and matters more than people expect: you own everything. Footage, project files, exports, all yours. If you ever want to take the whole library elsewhere, I will hand it over without a conversation about it.

If you ever record two people in the same room, or add a local camera to a remote call, Riverside handles the mixed setup more comfortably and gives an editor more to cut with.

A remote interview with a good local camera on the host plus a Riverside track for the guest gives me three usable angles once you count the wide. That turns a static two-box layout into something that feels like a show. Cutting between angles on natural beats is most of what makes multi-cam editing look expensive when it really just looks intentional.

Zencastr can absolutely be part of that setup. You record your local cameras separately and use Zencastr for the guest and for the reliable audio. It works, it just takes a little more coordination on your end, which is why I always ask for the clap.

One honest caveat about 4K. Recording at 4K is genuinely useful because it gives me room to reframe and crop for vertical without losing sharpness. It is also heavier to upload, heavier to store, and it will eat a guest's laptop if they are on an older machine. If your guest is on a five year old Windows laptop and a hotel wifi connection, 1080p is the responsible choice and the finished video will be perfectly good.

You drop the folder in a shared drive, we watch the entire episode, and the first cut comes back in 48 hours. Revisions come back the next working day.

Nothing is auto-cut by silence detection here. A human editor watches the whole thing, start to finish, which is how we catch the moment your guest went quiet because they were thinking rather than because the audio failed. Silence detection cannot tell those apart. It treats a dramatic pause and a dead mic as the same event, and it removes both.

The pass covers the same ground every time: remove the false starts and the tangents that go nowhere, level the audio so the quiet guest and the loud guest sit together, colour match the cameras so one person is not orange, sync everything to frame, then add your intro and lower thirds. If you want vertical clips they get cut from the same episode by the same person, so the clips reflect the same taste as the main cut.

One last piece that is outside my scope but affects your numbers more than any recorder choice: the thumbnail. A great episode with a weak thumbnail is a great episode nobody clicks. Our sister studio ClickTheory does YouTube thumbnails at 8 for $250, and hosts tell us that is the cheapest lever they pulled all year.

I am Samy. I am based in Dubai and I work with shows worldwide, entirely remote. Email me at samy@pixelsamy.com and tell me which platform you are on.