Should You Pitch Small Podcasts?

What catalogue size actually predicts

Three checks, run over every readable feed, split by how many episodes the show has published. Whether it is still publishing. Whether it offers a contact route beyond a bare homepage. Whether it publishes visible evidence that it books guests. Then all three at once, which is the list you could pitch today.

Read the columns against each other, because that is where the finding is. The share of shows publishing visible guest evidence barely moves: from 29% in one band to 33% in another, with no ladder to it. Catalogue size tells you almost nothing about whether a show books guests. What it tells you, and tells you loudly, is whether the show still exists.

Why the smallest shows look like the easiest ones

Most podcasts stop. They stop early, they stop quietly, and the feed stays up afterwards looking exactly like a feed that is still running. Nothing about a show announces that the last episode was the last one.

That is what our smallest band is mostly made of. 28 of 45 shows under the first cut had published nothing in six months when we read them. A show that quit after a dozen episodes is small forever, so the shortest catalogues collect the shows that never made it, and any list sorted by size sorts them straight to the top.

It also explains why the advice feels true while producing nothing. A small show is easier to reach, in the sense that there is no producer between you and the host, and it is genuinely likelier to be flattered by a good pitch. But a pitch to a show that stopped publishing last spring is not an easier booking. It is not a booking at all, and you will never find out why, because a dead show does not send rejections.

There is a second reason the smallest band underperforms, and it is not survivorship:

• Reachability is worse there too. 76% of the smallest catalogues publish a contact route beyond a homepage, against 93% of the largest. A show that has been running for years has usually been asked for an address enough times to publish one.
• Average openness runs the same way. On the zero to one hundred scale our directory ranks shows on, the smallest band averages 32 and the largest 54, against a corpus median of 46.
• The shows that pass every check are concentrated at the other end: 27 of them sit in the largest band alone, out of 70 in the whole corpus.

The catch with the biggest catalogues, stated plainly

This is the part the numbers above do not settle. Long-running shows are likelier to be network-run, and a network means a booking desk between your email and the host: 27% of the largest band are network shows, against 18% of the smallest. In our openness scoring, network shows average 35 and independents average 48, so the booking desk is a real cost and it lands hardest exactly where the qualified shows are.

So the useful cohort is not "big". It is long-running AND independent, and it is not a rare thing to find: 71 shows in our corpus have published 500 episodes or more with no network behind them, and they average 59 on the same scale, our highest-scoring group. A host who has published for a decade, answers their own mail, and has interviewed several hundred people is the single best target profile in this data.

The honest summary of the size question is that it is the wrong axis. Sorting by size is a proxy, and it is a proxy for two things at once that pull in opposite directions: longevity, which helps you, and institutional distance, which hurts you. Sort on the two directly and the proxy stops being needed.

What to do with your list instead

None of this means avoid small shows. It means stop treating small as a category of opportunity, because it is not one, and run the checks that actually separate a good target from a bad one. In order, cheapest first:

• Check the last episode date before anything else. It takes seconds and it removes more shows from a size-sorted list than any other check.
• Check that there is a contact route that is not a homepage contact form buried behind a menu. If a show has not published a way to reach it, a stranger has nowhere to start.
• Check whether it visibly books guests, by looking at recent episode titles. Ours is a conservative machine check, and yours will be better, because you can hear an interview a script cannot parse.
• Prefer independent over network at equal everything else, and prefer long-running over new at equal everything else. Those two beat size every time, and the second one is what the small-show advice was reaching for.
• Then spend the expensive twenty to forty minutes of research and writing only on what is left.

A small show that published last week, answers its own mail, and interviews someone every episode is an excellent pitch, and our data says nothing against it. A small show that stopped in the spring is the thing the advice keeps handing you, and it is most of what is down there.

Sources and benchmark date

Every figure above is derived at build time from our own read of the public feeds behind 360 podcasts between July 11, 2026 to August 16, 2026, not typed in. The method, the limitations and the raw data are public:

• Which Podcasts Actually Book Guests? (our study)

  Original analysis of the 338 readable feeds inside a 360-show directory. Source for the openness score, the network split and the definition of visible guest evidence used on this page. Method, limitations and the full dataset are published with it under CC BY 4.0.

• Is This Podcast Still Active? (the liveness check)

  The check that does the most work on this page, in detail: how often research hands you a show that has stopped publishing, and why the signals a directory sorts on do not decay when a show ends.

• Should You Pitch Podcasts on a Network?

  The control on the confound named above. Splitting the same corpus by whether a network runs the show, the size ladder still climbs among independents alone, so the finding here is not a network effect in disguise.

• How Many Podcasts Should You Pitch?

  The same three checks applied in sequence to the whole corpus rather than split by size: 70 of 360 shows passed all of them, which is what should size your list.

Frequently asked questions