The Podcast Index counted roughly 4.72 million podcast feeds in 2026, and about 323,000 of them published an episode in the previous thirty days. That means the overwhelming majority of shows that have ever launched are not currently making anything.
Editors know this. It is why “we are launching a podcast” lands on a desk with less force than founders expect, and why a well-written podcast press release that says only that will get no reply from anyone. The launch is not the story because launches are abundant and continuation is rare.
What follows is the structure that works instead, built around the four angles that actually get picked up, plus the template to put them in.
Nobody Covers a Launch
Start by accepting the premise, because every mistake downstream comes from resisting it.
A podcast launch has no news value on its own. It is a thing that is about to happen, announced by the person it benefits, with no outcome yet and no independent verification available. Those are the four properties editors use to identify something they can safely ignore.
Compare that with something a trade editor will run. A named person left a known company to do this. A show obtained something nobody else had. A category with no coverage is getting its first serious treatment, and here is the evidence that gap exists. Each of those has a subject other than you, which is the test.
This also reframes who you are writing for. A podcast press release is almost never read by a general assignment reporter. It is read by trade editors covering your subject area, by media and audio reporters covering the industry, and by newsletter writers who need something to point at this week. Those three audiences want different angles, and a single release written for all of them reaches none.
Lead With the Claim, Not the Date
Open the release with the most contestable true sentence you have.
Most launch releases open with the announcement. “Acme Industries today announced the launch of The Acme Podcast, a new show about logistics.” Every word there is predictable, and an editor skimming fifty emails has already moved on. The date is in the dateline. Repeating it in the first sentence wastes the only line you are guaranteed to get read.
Instead, open with the thing you are asserting. “Freight brokers are paying twice for the same load in roughly one of every nine transactions, according to a review of 400 settlements, and a new weekly show from Acme Industries starts with that number.” Now there is a claim, a source for it, and the show arrives as the vehicle rather than the subject.
The test for your opening line is whether a reader who disagrees with you could say so. If nobody could disagree with your first sentence, it is not a claim, it is an announcement, and announcements do not travel.
Three practical constraints on that line. It has to be true and checkable, because a trade editor will check it and a failed check ends the relationship rather than the story. It has to be specific enough to cite, which usually means a number with a method behind it. And it has to be yours, meaning the show produced it rather than repeating a statistic from somewhere else.
The Launch-Window Grid
Here is the model we use to pick the angle before writing a word, and it has four cells. Call it the Launch-Window Grid. Every podcast launch has at least one of these and almost none have all four, so the work is identifying which one you genuinely hold.
The first cell is the person. Somebody recognisable in the category is behind the show, and their move is the story. This works when the name carries weight with the specific publication you are pitching, and it fails badly when you overestimate that weight. The signal is whether the person has been quoted by that outlet before.
The second cell is the access. The show has a guest, a dataset or a location nobody else has secured, and the first episode proves it. A first interview with someone who has not spoken publicly is a genuine story and will be covered as the interview rather than as the podcast, which is the correct outcome. Do not bury this one. If episode one has real access, the access is the entire release.
The third cell is the gap. A subject with commercial or civic weight has no sustained coverage, and you can demonstrate the absence rather than assert it. This is the most available angle for most shows and the least used, because demonstrating a gap takes work. Counting how many shows in a category published in the last ninety days, using the public index data, is a defensible way to do it.
The fourth cell is the format. The show does something structurally unusual that other producers would find notable. This is a narrow angle that works almost exclusively with audio and media trade press, and it is worthless to subject-matter editors who do not care how the sausage is made.
Pick one cell. Write the release for it. If you hold two, write two releases for two audiences rather than one release that mentions both.
The Template
Use this structure and resist decorating it.
The headline states the claim or the access in under fifteen words, with no colon and no show tagline. “Freight brokers double-pay on one in nine loads, new Acme data shows” beats anything containing the word “launches.”
The dateline carries the city and the date, and that is where the date lives.
The first paragraph gives the claim, the evidence for it in a clause, and the show’s existence in the final clause. Three sentences maximum. An editor should be able to stop here and know whether they want it.
The second paragraph gives the method. How many records, over what period, defined how. This paragraph is what separates a release that gets checked and run from one that gets deleted, and most releases skip it entirely.
The third paragraph is the quote, and it has one job: to say something the data cannot. Quotes that restate the first paragraph in warmer language are the most common filler in the format. Have the host say what surprised them, or what they expected to find and did not.
The fourth paragraph covers the show itself. Name, host, cadence, episode length, where it is available, and the first three episode subjects as a plain list in prose. This is the only paragraph where promotional detail belongs, and it belongs here because by now the editor has a reason to care.
The fifth paragraph is the boilerplate, kept to three sentences about the organisation.
Then the contact block, with a human name, a direct email and a phone number that someone will answer. An unattended press address at the bottom of a podcast press release is a quiet way of saying nobody is really expecting a call.
Below the contact block, list what is available on request: the full dataset, an advance copy of episode one, the host’s availability for interview with specific windows, and a photograph at print resolution. Editors respond to specificity here more than to anything in the body.
Build the Evidence Before You Need It
The reason most shows cannot write a claim-led release is that they have no claim, and the reason for that is sequencing. The release gets written in the week before launch, by which point the only available facts are the show’s own existence.
Fix that by producing the evidence during pre-production rather than after. If you are recording six episodes before launch, you are already gathering material that nobody has assembled: the same question asked of six practitioners, the point where their answers diverge, the thing four of them said off the record that three will say on it. That is a finding, and it costs nothing extra because the recording was happening anyway.
The cheapest version is a count. Pick the question your category argues about, ask every guest the same version of it, and report the distribution. Six answers is not a survey and you should not call it one, but “five of the first six supply chain directors we interviewed said the same thing, and the sixth explained why they are all wrong” is a real sentence that a trade editor can run.
The second cheapest is a document. Public filings, inspection records, rate sheets and settlement data exist in most commercial categories and almost nobody reads them. Reading two hundred of anything gives you a number, and a number with a stated method is the raw material of a podcast press release that works.
Do this four to six weeks out. By launch week you want to be choosing between two claims rather than inventing one.
Send It to People, Not to a Wire
Wire distribution has a narrow legitimate use and launch announcements are not it.
A wire puts your release on syndication sites where it will be indexed and almost never read by a journalist who was not already looking. For a publicly traded company with disclosure obligations, or for a release that needs a timestamped record of public dissemination, that has value. For a show launch, it produces a set of low-quality links and a report you can show your board.
The alternative takes longer and works. Build a list of fifteen to twenty-five named people: trade editors who cover your subject, reporters who have written about your category in the last six months, and newsletter writers in the space. Read something each of them wrote. Send individual emails with the claim in the subject line and the release pasted below the signature rather than attached.
Twenty-five real emails sent on a Tuesday morning will outperform a wire release every time, and the follow-up is manageable. One follow-up after four working days, then stop.
Timing matters less than people believe, with one exception. If your claim touches a subject that already has a news cycle running, send into it immediately rather than waiting for your planned launch date. The release is a vehicle for the claim, and claims expire.
Where These Go Wrong
Four failures account for nearly all of it.
The release is about the show rather than about something the show found. This is the structural error and no amount of polish fixes it.
The claim is not checkable. A number with no method behind it reads as marketing to anyone whose job is verification, and trade editors are the most sceptical readers you will ever have because they know the category.
The list is wrong. Pitching a general business desk on an audio format story, or an audio trade publication on a subject-matter claim, wastes the one send you get with each of them.
And the first episode does not deliver what the release promised. This is the only mistake with lasting cost. An editor who covers your claim and then hears an episode that does not support it will not read your next release, and the trade press in any category is small enough that this travels.
So before you send anything, listen to episode one as though you were the editor who ran the story. Does it hold up the sentence you put at the top?