What makes a good podcast

Most podcasts do not fail because they were bad. They stop. Somewhere around episode seven the editing backlog and the quiet download numbers and the general question of whether anyone is out there add up, and the next one never gets recorded.

So the first quality of a good podcast is unglamorous: it exists repeatedly. Everything below is in service of that, because a show that is sustainable gets the chance to get good, and a show that is perfect once does not.

Which means the real question is not how to make a good episode. It is how to make one you could make again next week without dreading it.

The shows people describe as good almost always feel addressed. There is a sense that the host knows who is listening and is talking to them rather than broadcasting past them.

You get that by narrowing, which feels like losing audience and is the opposite. A show for anyone interested in business is competing with everything. A show for people who left a corporate job to open a practice on their own has listeners who feel personally seen, and those listeners tell other people.

A useful test: can you name one real person, by name, who would put this in their queue? If not, the show is still aimed at a category rather than a human.

Good episodes have a shape you can sense without being told. Something opens, something develops, something closes. The listener relaxes because they can feel where they are in it.

This is also the thing that makes a show repeatable. A format you can fill is a format you can do on a week when you are tired, which is most weeks. Structure is what carries you when inspiration does not.

It does not need to be complicated. A question, an exploration, a conclusion is enough. What matters is that it is the same enough each time that your listener knows the rhythm.

Audio matters in one direction. Bad sound makes people leave. Good sound does not make them stay. Nobody has ever recommended a podcast on the basis of its noise floor.

The floor is lower than most beginners think and higher than a laptop mic in a kitchen. A directional microphone, four to six inches away, a room with something soft in it, consistent levels. That is the whole bar, and it is reachable in an afternoon.

Past that point, extra production effort mostly buys diminishing returns and a heavier workflow. Which brings you back to step one, because the heavier the workflow, the sooner it stops.

New hosts often adopt a register that is not theirs. Slightly lower, slightly more formal, a bit of radio in it. It is a reasonable instinct and it is the fastest way to sound like everyone else.

Listeners choose podcasts for the company. They are letting a person talk in their ear while they drive or cook or walk. What holds them is the sense of a specific mind, and presenter voice sands that off.

If you find your recorded voice uncomfortable, that is normal and it fades. It is not a reason to perform someone else. Your ordinary speaking voice is the asset.

Every show worth listening to believes something. Not a brand position, an actual view the host would defend. That is what makes an episode feel like it was made rather than assembled.

You cannot manufacture that in the edit. You can only give it enough episodes to come through, which means the whole thing loops back to whether the process is light enough to keep doing. Make it easy to make. Then have something to say. In that order.