Which podcasts are worth your time, and how to vet them

An invitation arrives, or a show reaches the top of your list. Before you answer, the question is whether this one is worth half a day.

That is what an appearance costs, once you count listening, preparation, the recording, and the follow-through. Shows differ enormously, and the difference between a good yes and a bad one is checkable in an hour.

Judge the audience before the audience size

The right show is the one whose listeners could hire you. That sounds obvious until you notice how much guesting advice ranks shows by reach instead.

A room of 200 buyers beats a room of 10,000 peers. The peers will enjoy the episode. They will not call you.

You can read the audience from public signals. The last ten episode titles tell you what the show is for. The guest list tells you who the host thinks her listeners respect. The sponsors, if there are any, tell you which buyers the show believes it reaches. Somebody paid for that belief.

Then ask the host directly. Who listens to this show, and what do they do for work? A host who knows her audience answers in a sentence. A host who can only describe an age range is telling you something.

Download numbers are self-reported, and the bar is lower than it sounds

No independent body audits a show's audience. What a host tells you is what her dashboard says, or what she remembers it saying.

So the useful move is to know what the numbers mean. Buzzsprout publishes platform-wide figures every month across its active shows. In July 2026, the median episode drew 27 downloads in its first seven days. An episode at 97 was in the top quarter, and one at 409 was in the top tenth.

A show at 500 downloads an episode sits near the top of the field, and it is still a room of 500 people.

That reframes the conversation. A host who mentions a few thousand downloads is running a serious show, or repeating a number nobody checked. Both happen.

Ask for a screenshot from her hosting dashboard. Serious hosts send one without friction. They are proud of the number, and they know the question is fair.

Percentile brags describe a directory, not an audience

"Top 2% globally" appears on a great many show pages. It usually comes from Listen Notes, whose global rank orders more than 3.7 million feeds.

Most feeds in a directory that size are dormant. Being ahead of them is a low bar. The score is relative by design, which the company says on the same page.

Apple category charts have a similar problem. They respond to recent subscriber movement rather than audience size. A show can chart for a week on a small burst, then quote it for years.

None of this makes a host dishonest. She is repeating the most flattering true thing she has. Your job is to convert it into a number of humans, which takes one question.

Check that the show is alive

Dormancy is the most common condition in podcasting. About 4.8 million feeds were indexed in August 2026. Roughly 600,000 had published anything in the previous 90 days.

Four checks take five minutes each. Has it published in the last month, on something like a schedule. Does it run guest interviews as its normal format, rather than as an occasional experiment. Do its episodes carry reviews with sentences in them, rather than a wall of ratings and nothing else. Does each episode get a page with show notes and a transcript.

That last one matters more than it looks. The episode page is where your appearance lives after the week it airs. What happens to an episode afterward is where most of the return sits.

If the show posts full episodes to YouTube, you get something rare: a public view count per episode. Compare it against what the host told you. The two numbers measure different things, so they will not match, but the gap is informative.

Price the hour before you spend it

Listening to two episodes and reading the show notes takes an hour. Preparation runs one to three hours for most people. The recording runs 45 to 75 minutes. Then the clips, the post, the page.

Half a day is the standard price, and travel to an in-person recording doubles it.

Set that against what you are buying. Half a day for 200 of the right listeners is a good trade at any stage of a practice. Half a day for 40 listeners who will never hire anyone is not. No amount of preparation improves it.

The arithmetic also explains why volume advice misleads. A dozen well-chosen appearances is six days of work. Fifty appearances is a part-time job you did not agree to.

When a small show deserves a yes anyway

Three cases justify saying yes to a show with a modest audience.

The audience is exactly yours. Forty listeners who all run manufacturing firms in your state beat 4,000 general business listeners.

The host is connected. Hosts talk to other hosts, and a good appearance produces the introduction that gets you the next three.

You need the recording. If nobody can hear you talk yet, an early episode becomes the sample every later host will check. That check decides bookings. Treat your first few as training, and choose them for the practice rather than the reach.

Decline in a way you can come back from

Guesting advice rarely covers the no, which is strange, because you will send more declines than pitches.

Keep it short and specific. Thank her, name the reason in plain terms, and leave the door open. Something like: my audience is HR directors, and I do not think I would serve your listeners well. I would like to stay in touch as the show grows.

Two things to avoid. Do not ghost an invitation, since hosts talk and the medium is small. Do not accept out of politeness. A guest who did not want to be there is audible, and that episode becomes somebody else's sample.

If you like the host but not the fit, refer someone. It costs you an email and it makes you the person she remembers.

Take the last invitation you accepted and run two checks on it. Could its listeners hire you, and do you know how many of them there are. If either answer is no, you have learned what your yes has been worth. The next one gets easier.