Video podcast vs audio podcast: which should you make?

Framed as a question about formats, this is hard to answer. Framed as a question about one recording session, it becomes much easier, because you are not really choosing between two products. You are choosing whether there is a camera running while you talk, and everything downstream follows from that single decision.

What follows is the honest version, including the parts of video production that cost you something real. We build both kinds of show, and the base edit costs the same either way, so there is nothing to sell you here in one direction or the other.

What is the difference between a video podcast and an audio podcast?

An audio podcast is a conversation recorded to audio and distributed through podcast feeds, reaching listeners through Apple Podcasts, Spotify and any other app that reads an RSS feed. A video podcast is the same conversation captured on camera as well, published on video platforms alongside those feeds, and usually cut so it works when watched rather than only when heard.

The important part is that these are two outputs of one act. A video podcast contains an audio podcast inside it. The reverse is not true, and that asymmetry is the whole argument. Deciding to shoot video is reversible at any point later, because you can always publish audio only. Deciding not to shoot video is not reversible: no amount of editing recovers a picture that was never recorded.

So the practical question is not "which format is better". It is "is the additional cost of running cameras worth what the picture makes possible", and that depends on your show more than on any general trend.

What does each format cost you?

Less in editing than most people assume, and more in the parts nobody mentions.

Setup and recording discipline. Audio forgives a lot. You can record a good audio episode in a soft-furnished room with two decent microphones and no thought given to what anything looks like. Video does not forgive: framing, eyelines, backgrounds, lighting that does not change halfway through, and the discipline of not leaning out of shot. That is a real ongoing tax on the host's attention during a conversation where attention is the main resource.

File weight, which is the underrated one. A single episode of one three-camera show we work with arrives as three camera masters at roughly 610 to 665MB each, plus two combined audio tracks at about 67MB. On a larger rig, per-angle masters for one episode have run anywhere from roughly 10GB to 56GB. An audio-only handover for the same conversation is a couple of hundred megabytes at most. This changes your week in concrete ways: how long files take to upload, how much drive space you need, how quickly an editor can even begin. Shows that record on Friday and want a cut on Monday routinely forget that the files have to physically travel first.

A failure mode that belongs specifically to video and multitrack recording. On one remote interview show, the guest's answers were largely missing from the finished transcript. Not because of anything the guest did, but because each speaker recorded to their own separate file and the guest's track never made it into the multicam sequence. The audio had existed the entire time, sitting on a machine in another country. Audio-only shows recorded to a single mixed file almost never break this way. It is a genuine cost of the video workflow, and no equipment guide mentions it because it is not an equipment problem.

Post-production, which is where the surprise is. Every extra angle means another sync, another set of switch decisions, another thing that can drift out of frame. But at Joycast the base edit is £280 per episode whether the show is video led or audio led, covering up to an hour of finished episode. The cost difference between formats lands in your recording setup and your own time, not in the editing invoice. That is worth knowing before someone quotes you a video premium.

What does each format earn you?

Audio earns you attention during other activities. People listen while commuting, walking, cooking and working. That is a genuinely privileged position and video does not replace it. Long, discursive, one-thing-at-a-time conversations often work better as audio, because nobody is watching anyway and the picture is just overhead.

Video earns you two things. The first is a second place to be found, with its own search behaviour and its own recommendations, entirely separate from podcast apps. The second, and in our experience the more valuable one, is raw material.

You cannot cut a talking-head vertical clip out of an audio-only recording. Everything in the short-form pipeline, the clips that actually introduce strangers to a show, depends on a camera having been running. In practice this is the real reason most of our clients shoot video, and plenty of them are honest that very few people watch the full episode on a video platform. The full video is not the point. The forty clips a year are the point.

Some shows genuinely earn their video, though. One client of ours is a workshop where objects are made by hand, and the camera is doing real work there: you can see the thing being made, and the picture carries information the audio cannot. If your subject is visual, video is not a repurposing strategy, it is the product.

Most shows are not that. Most shows are two intelligent people in chairs, where the video's honest job is to feed clips and hold a presence on a platform your audience already uses. That is still a good reason. It is just worth being clear-eyed that it is the reason, rather than expecting an hour of chairs to be compelling television.

Should you do both from one recording?

For most shows, yes, and this is the practical answer the format debate usually misses.

Record video. Publish the audio to your feeds exactly as you would have anyway. Publish the video where video lives. Cut clips from the video. One recording, three outputs, and the marginal cost of the extra two is small once the camera is already running.

What that requires in practice is modest: record every speaker to their own track as well as the cameras, keep the local recordings rather than relying only on what a platform streams, and check that every file exists before anyone leaves the call. That last one takes thirty seconds and prevents the missing-guest-track problem described above. The full setup, including what to buy and what genuinely does not matter, is in our video podcast production guide.

The case for audio only is narrower but real: you hate being on camera and it visibly affects how you speak, your show is genuinely audio native such as narrative or fiction, or you are testing whether the show survives ten episodes before spending anything on a room. All three are legitimate. The first is the one people are least honest with themselves about, and a host who is uncomfortable on camera makes a worse show on camera, which is a fair trade against the clips.

Where do you go from here?

Answer one question honestly: if nobody watched the full video, would the clips alone justify running a camera? For most shows the answer is yes, and that settles it without any need to predict where podcasting is heading.

If you would like both formats out of one recording, spec an episode on the configurator and see what a video edit with clips and artwork actually costs, or read what the edit itself includes on the editing service page. And if audio only is right for your show, do that properly instead of shooting reluctant video badly. A well-made audio show beats an uncomfortable video one in every format that matters.