There are now more podcasts than any listener could work through in several lifetimes. Most of them aren't bad — they're simply unnecessary. They exist because someone decided the brand should have a podcast, without asking the harder question: what would make this worth somebody's time and attention? That question is where every great podcast starts.
And it's an uncomfortable question, because the honest answer is almost always «nothing, not yet». Nobody is out there hoping for one more interview show in which the host asks the guest about their journey. The attention a podcast asks for is enormous: forty uninterrupted minutes, ears occupied, no option to skim. A badly written article is abandoned in ten seconds. A boring episode takes half an hour before the person realises it was taken — and they don't come back.
What separates the shows that earn devoted listeners from the ones nobody opens twice comes down, nearly always, to three things: having something to say, sounding good saying it, and having the discipline to cut everything else.
The gift of a genuine perspective
Podcasts that build loyal audiences have a point of view. They don't try to be everything to everyone. They say: this is what we believe, this is the lens we see the world through, and we're going to share it consistently and with conviction. That specificity is what turns casual listeners into devoted ones.
There's a simple test for whether you have a perspective or merely a topic: could someone competent and well-informed disagree with you? If your position is that sustainability matters, that team culture counts, or that the client should be at the centre, you don't have a position — you have consensus, and consensus doesn't get listened to because it's already known. If your position is that most small companies shouldn't seek investment, or that remote working destroys more than it solves in creative teams, some people will frown. Those people listen to the end.
Perspective shows up in the questions
Having a point of view doesn't mean monologuing. In an interview show, the host's perspective lives in the questions they ask and, above all, in the ones they refuse to ask.
A show without an angle asks the guest how they got started, what the challenges in the sector are, and what advice they'd give to someone beginning. Those are questions the guest has been asked twenty times and answers on autopilot, with the same three sentences they used at the last conference. A show with an angle asks what went wrong on that specific project, why they made the decision everyone thought was wrong, what it actually cost. The difference isn't aggression — it's preparation, and it's having a position.
Guest selection communicates the same thing. If you only invite comfortable people, from companies that look like yours, who say things that validate you, the show becomes a mirror. Inviting someone who publicly disagrees with you is one of the best episodes you can make, as long as you invite them to argue and not to be ambushed.
Specificity instead of breadth
The temptation for anyone making a brand podcast is to broaden: to talk about everything that might interest everyone who might one day buy something. The effect is the opposite of the one intended. A show that serves a few hundred of the right people precisely is worth infinitely more than a show that annoys nobody.
That means accepting episode titles that put off most listeners. «How to manage a team» persuades nobody to press play. «What to do when you have to let go of someone you like» persuades exactly the people who are going through it — who are the ones who'll listen to the end and tell someone else about the show.
Sound design isn't optional
Audio quality is the bare minimum. A podcast recorded in a room with bad acoustics, with background noise, with inconsistent levels between speakers, communicates carelessness no matter how good the content is. And it communicates it before the first sentence is over: everyone has listened to radio and television their whole life, and everyone recognises amateur sound within two seconds, even without being able to name what's wrong.
What ruins a recording
Most sound problems aren't solved by more expensive equipment. They're solved by decisions taken before anyone presses the red button.
- The room. A space with bare walls and a glass table throws the sound back and produces that corridor echo no edit cleans up properly. Curtains, a rug, full bookshelves, a sofa — any soft surface improves the recording more than changing microphone.
- The distance. The most common mistake is having the microphone too far away. A hand's width from the mouth, the voice has body and the room noise disappears. At half a metre, it's the reverse.
- Continuous noise. The fridge, air conditioning, a laptop fan, traffic through a window someone left open. The human ear ignores these live and stops ignoring them in the recording.
- Body noise. The chair that creaks, rings against the table, pages being turned, legs knocking the tripod. It's worth warning everyone recording, guests included, before you start.
- Remote recording. Never trust the call audio. Each person records locally, and the tracks are brought together afterwards. A weak connection irreversibly ruins a conversation you can't repeat.
Recording on separate tracks, even with everyone in the same room, is one of the highest-return decisions you can make. It lets you fix uneven levels, delete a cough that lands over someone else's answer, and cut without dragging the other voices along with it.
Sonic identity is brand identity
Once the technical side is settled, the sound design proper begins: the music, the transitions and the overall sonic character of the show. These are the elements that make the experience coherent and memorable — and they're also what separates a show from a recording of a meeting.
The opening is where most of this is decided. Fifteen recognisable seconds do the same job as a logo: they tell the person, before they've even looked at their phone, which show is starting. A stock library track chosen in a hurry performs the function generically — and generic is precisely what a brand show can't afford. A short theme, composed or at least edited to measure, with a signature that recurs at the close and in the transitions, is worth the investment because it serves dozens of episodes.
The same principle applies as with typography or motion on a website: the character has to match the brand. A financial consultancy with a cheerful guitar theme sounds false. A children's products brand with solemn strings sounds like somebody else's company. And whatever the choice, music underneath a voice has to be low enough that it's almost felt rather than heard — it's one of the most frequent mistakes in shows that want to sound produced.
The podcast is made in the edit
An hour-long conversation rarely makes an hour-long podcast. The edit is where rambling becomes clarity, where pace is controlled, where the best forty minutes of a ninety-minute conversation are preserved and everything else is removed. Good editing is invisible. It makes the listener feel the conversation was always that sharp.
What gets cut
The first ten minutes, nearly always. People need to warm up, and a guest's first answer is usually the brochure version of their second, which is the good one. The questions that led nowhere, including the ones you worked hardest to prepare. The explanations the guest already gave better five minutes earlier. The in-jokes that only land for people who were in the room.
And you cut the verbal tics — «so», «right», «you know», long hesitations — but with restraint. Stripping out every breath and pause produces something strange and airless that sounds like machine-gun fire and tires the listener more than the original rambling did. The silence after a strong answer is an instrument: it gives the listener time to register what they just heard.
Reordering is legitimate
The edit isn't obliged to follow the chronology of the recording. If the best thing anyone said arrived at minute fifty, that sentence can open the episode — a few seconds before the theme, as a taster — and then reappear in its place. Whole blocks can change order if that makes the argument easier to follow.
The limit is ethical and it's clear: you can improve the form, never the content. Joining two separate answers to make it sound like someone said something they didn't, or cutting the qualification that came right after the strong statement, isn't editing — it's falsification. When in doubt, send the cut to the guest before publishing.
Listen the way people listen
Before you finish an episode, listen to it outside the studio: in the car, walking down the street, on cheap earphones while you do something else. That's how your audience will hear it, and that's where the problems show up — the section where attention drifts, the voice that sits too low against street noise, the two minutes that could be cut whole without anyone missing them.
What's at stake
A brand podcast that wins genuine listeners is an extraordinary asset. It puts your thinking directly into someone's ears while they commute, exercise or cook. That level of intimacy and sustained attention is something no other medium replicates.
But it's intimacy granted, not taken by force, and it's withdrawn as quickly as it's given. It demands the same respect for the audience's time that any creative work worth doing demands: having a position, sounding like you mean it, and cutting everything that only interests the person who said it.