It is Tuesday night. You have a spreadsheet open with forty podcast names on it, sorted by a download estimate you pulled off a rankings site. One pitch is written and pasted into thirty-eight of the rows. The subject line reads “Podcast guest inquiry.” You are about to send the batch, and some part of you already knows how this goes, because you ran the same play in March and collected two auto-replies, one unsubscribe, and a single polite no.
The problem is not your credentials. It is that you are pitching a slot when the host is buying an episode. Those are different products. A slot is a hole in a calendar. An episode is forty minutes of audio that has to hold a commute, survive a chapter list, and give the host something to clip for social. Every rule below exists to close the gap between what you are offering and what the host is actually shopping for. If you want to pitch a podcast as a guest and get booked, you have to think like the person who has to publish the thing.
Pick shows by overlap, not by download count
Most people who set out to pitch a podcast as a guest start by ranking shows, and they rank them wrong. Download numbers are the worst input available to you, and they are the one everybody uses. Public estimates for podcast audiences are guesses built on chart position and review counts. Even accurate ones tell you nothing about whether the people listening are the people you need.
Sort your list by audience overlap instead. The question is not how many people hear the show. It is what those people do for work, what they buy, and whether an hour of you talking would change anything for them. A show with four thousand engaged listeners who all run operations at mid-market manufacturers is worth more to a supply chain software founder than a general business show with two hundred thousand.
There is a second filter most people skip: format fit. Some shows are guest-driven and run an intake pipeline. Others are host-driven and built around research, where guests appear because the hosts chose to cover their company. Acquired, the show Ben Gilbert and David Rosenthal built around multi-hour company histories, is the clearest example of the second kind. You do not pitch your way onto an episode about a company you do not run. Understanding which kind of show you are looking at saves you from sending pitches that were never going to work, no matter how good they were.
Build the list in three tiers. Tier one is five shows where your topic is the show’s topic and the host takes pitches. Tier two is fifteen adjacent shows where you would be an interesting outsider. Tier three is everything else, and you should probably delete it. Forty names sorted by downloads is a worse asset than twenty names sorted by fit.
One more input worth checking before a show goes on the list: publishing consistency. A podcast that shipped weekly for two years and then went quiet for four months is either between seasons or dead, and the difference matters. Look at the release dates on the last ten episodes. If the gaps are widening, the host is losing steam, and a booking on a show that stops publishing in November is worth nothing. Shows that publish on a rigid schedule tend to have a queue, which means a slower yes but a real one.
Check the guest history too. Scroll a year of episode titles. If every guest is a founder at a company larger than yours by an order of magnitude, you are not the profile, and the polite version of that fact will arrive as silence. If the guest list is a mix, including people you have never heard of, the host is casting for conversation rather than for logos, and that is a show you can get on.
Listen to three episodes before you write a word
This is the step everyone claims to do and almost nobody does. Three episodes, at speed if you must, with a notes file open.
You are listening for four things. First, the show’s actual question, which is rarely the one in the description. Some shows are about tactics. Some are about origin stories. Some are about the host working out an idea in public and using guests as sounding boards. Second, the guest profile: are they operators, academics, authors, or founders, and at what stage. Third, the segments the host uses, because if there is a recurring closing question or a rapid-fire round, you can answer it in your pitch and prove you were there. Fourth, what the host is tired of. Hosts telegraph this. They interrupt, they redirect, they say some version of “everybody says that.”
That fourth item is the one that books you. If you heard the host push back twice on the standard answer about hiring, and you have a genuinely different position on hiring, you now have a pitch. Not a bio. A pitch.
Take one specific detail from each episode you heard and keep it. You will use exactly one of them, and it will be the line that gets you a reply. Referencing a specific moment from a specific episode does two things at once: it proves you listened, and it tells the host what you found interesting about their work, which is information they want.
Write the cold open pitch
Here is the framework. Call it the Cold Open Pitch, and build every guest email around it.
A podcast cold open is the twenty seconds of tape that runs before the theme music. It is a clip, pulled from the middle of the interview, chosen because it makes a stranger stop scrolling. It is never a bio. It is never “today I’m joined by.” It is one specific, slightly uncomfortable claim, stated by the guest, without setup.
Write your pitch the same way. The first line of your email is your cold open: one sentence, no throat-clearing, that states the claim you would make on the show. Not your title. Not your company. The claim.
Weak: “I’m the founder of a logistics software company and would love to come on the show to discuss supply chain trends.” Strong: “Most warehouse automation projects fail in the first ninety days, and it is almost never the robots.” The second one is a cold open. The host can hear it in their feed.
Then three short paragraphs. Paragraph one: two sentences of evidence that you can defend the claim, using what you have done rather than what you are called. Paragraph two: the specific moment in their show that made you write, and how your claim extends or contradicts it. Paragraph three: three angles you could take, one line each, written as episode topics rather than as things you know about. End with a single question that makes yes easy.
Nothing else goes in. No attached media kit. No headshot. No list of other shows you have been on, unless one of them is a show this host would recognize and respect, in which case it is a half-sentence. Your one-sheet is a liability in the first email; it signals that you are running a campaign, and hosts can smell a campaign.
The whole thing is under a hundred and eighty words. Read it aloud. If it sounds like a person who listens to the show, send it. If it sounds like a press release with a name at the bottom, start over.
Hand the host a segment, not a bio
Hosts are not evaluating you. They are evaluating an episode they have not made yet, and they are doing it in about eleven seconds. Your job is to make the episode easy to picture.
That means angles, and angles are specific. “The future of work” is not an angle. “Why we moved a forty-person team back to four days in the office and what it cost us” is an angle. It has a claim, a number, a decision, and an implied conflict. A host can see the chapter list.
Give three. Three is the right number because it lets the host pick, which converts your email from a request into a choice. It also demonstrates range without making you look scattered, and it hedges against the possibility that your favorite angle is one the show covered four months ago.
Each angle should carry proof you can only get from having done the thing. Numbers you own. A decision you regret. A process you built and then abandoned. The material that makes good audio is the material that costs you something to say. Hosts know the difference between a guest who has a framework and a guest who has scar tissue, and they book the second one, because the second one produces clips.
There is a version of this that goes wrong, and it is worth naming. Some people read “be specific” as “be provocative,” and they build an angle around a contrarian take they do not actually hold. Hosts catch this in the first ten minutes of tape, and the episode gets shorter, flatter, and sometimes shelved. The angle has to be a position you would defend at dinner with someone who disagrees. If you find yourself constructing one for the pitch, you are not ready to pitch that show yet.
One more thing to include, briefly: what you will not talk about. If there is a lawsuit, a former employer, or a client you cannot name, say so in one clause. Hosts have been burned by guests who agreed to everything and then went vague on tape. Naming your fence early makes you look like a professional rather than a risk.
Send it to a person, not to a form
Guest submission forms exist to absorb volume. Some shows genuinely staff them. Most treat them as a spam filter with a nice interface. If the show has a producer, the producer is the person who books, and the producer has a name.
Find it in the show notes, in the credits read at the end of episodes, on the show’s site, or on LinkedIn under the show’s name. If there is no producer, the host handles booking, and the host has an email somewhere: a newsletter footer, a personal site, a Substack. Use the address they use for the show, not one you scraped.
When you truly cannot find a person, use the form, and then send the same pitch as a short note on whichever platform the host is active on. Not a pitch in the DM. A two-line note saying you sent something through the form and here is the one-sentence version. That combination gets read.
Timing matters less than people claim, but it is not nothing. Tuesday through Thursday, morning in the host’s time zone, avoiding the week a show is releasing a big series. If the host just announced a season break, wait. If they just put out a call for guests on a specific topic and you fit it, drop everything and answer within the day, because those windows close in hours.
Follow up once, then make the next slot easier
Everyone who has ever tried to pitch a podcast as a guest gets the follow-up wrong in the same direction: too many, too soon, too apologetic. One follow-up, seven to ten days out, three sentences. Restate the claim, add one new piece of information that did not exist when you first wrote, and give them an out. That is it. The second follow-up costs you the relationship and you will need that relationship in eighteen months when your story is better.
If they pass, ask one question: is there a topic you would come back for. Hosts answer that question honestly because it costs them nothing, and the answer is a booking brief for next year.
If they book you, the work is not done. Show up having listened to two more episodes. Bring three stories with numbers in them. Do not read from notes. After it airs, share it once, tag the show, and send the host a note that says which part you thought worked. Hosts get almost no feedback on their own work, and a guest who gives them some becomes a guest they recommend to the three other hosts they talk to every week.
Your next step tonight: open your spreadsheet, delete every row you cannot justify with one sentence about audience overlap, and pick the single show at the top. Listen to three episodes this week. Write one Cold Open Pitch on Friday. Send it Tuesday morning. That one email will outperform the thirty-eight you were about to send, and you will know within a week whether your claim is any good.