How to Monetize a Podcast With Sponsors
The money is priced per thousand downloads, so the number you can prove is worth more than the number you have.
You monetize a podcast with sponsors on a CPM basis: they pay a price per thousand downloads of the episode carrying their ad. You are not paid for subscribers, followers or reviews. You are paid for a measured number.
Current US rates sit around eighteen to twenty-five dollars per thousand for a pre-roll spot and twenty-five to thirty for a mid-roll, with host-read ads commanding the top of each band.
So a show doing five thousand downloads an episode with one mid-roll is looking at roughly a hundred and thirty dollars per episode. The arithmetic is unglamorous and it is the whole business.
Which is why the first thing a sponsor asks for is not your audience. It is your measurement.
The detail that decides whether a sponsor takes you seriously is how your downloads are counted, and it is the part most shows never think about until a deal stalls.
A download, at the server level, is a request for your audio file. That request can come from someone who listened to the whole episode, someone who pressed play and quit after four seconds, or an app that automatically fetched it overnight for a user who never opened it. In a raw log, all three look the same.
The IAB's podcast measurement guidelines exist to make them look different. Certified counting requires a minimum number of bytes delivered before a request counts, deduplicates by listener and day, and filters known bots. The result is a smaller number than your raw log, often much smaller.
That smaller number is the one worth more money. A sponsor comparing a certified eight thousand against an uncertified twelve thousand will buy the eight thousand, because they can model what it returns.
If your host is not certified, the practical move is to say so plainly and quote the raw number as raw. Sponsors have seen inflated figures before; what ends a conversation is discovering the inflation themselves.
The three slots and what each is worth
An episode has three advertising positions, and they are not priced alike.
Pre-roll runs in the first minute. It reaches everyone who pressed play, including the people who will quit, so it has the widest reach and the lowest intent. It is the cheapest of the three.
Mid-roll runs somewhere in the middle, usually at a natural break. Anyone still listening has already committed fifteen or twenty minutes, so the audience is smaller and considerably more valuable. This is the slot that commands the highest rate and the one sponsors ask for first.
Post-roll runs after the content ends. Most listeners are gone. It is sold at a discount and is often thrown in as part of a package rather than bought on its own.
A show that sells all three on every episode burns audience goodwill fast. Most sustainable shows run one mid-roll and treat the other two as inventory they can open when a deal justifies it.
Host-read against the ad somebody inserts for you
A host-read spot is you, in your own voice, in the recording. A programmatic spot is audio stitched into the file at delivery, which means different listeners hear different ads, and ads can be swapped into old episodes long after publication.
Host-read pays more per thousand because it converts better. The listener chose your voice; an ad in that voice does not read as an interruption in the same way.
Programmatic pays less but fills inventory you could never sell yourself, works on the back catalog, and requires nothing from you once it is set up. For a show without a sales operation, it is the realistic floor.
The pragmatic arrangement for a mid-sized show is both: host-read for the sponsors you found yourself, programmatic filling whatever is left. The mistake is treating programmatic as the goal rather than the baseline, because the rate difference between the two is roughly double.
The download number networks ask for
Advertising networks publish thresholds, and the figure that comes up most often is five thousand downloads per episode within the first thirty days.
Some take shows lower, particularly in niches sponsors find hard to reach otherwise — a two thousand download show about commercial insurance is more sellable than a ten thousand download comedy show, because the audience is identifiable and expensive to reach anywhere else.
Below that, networks are not the route. Direct sales are, and so is affiliate.
Thirty days matters as much as the number. Podcast listening is long-tailed, so a show can accumulate a respectable lifetime figure while having a weak thirty-day window. Sponsors buy the window, because that is when their campaign is running.
Selling direct before anyone will represent you
Direct sales are how small shows get paid, and the approach that works is narrow.
Start with companies your audience already buys from. Not the biggest advertiser in your category — the one whose customers look exactly like your listeners, and who is small enough that a marketing person will read your email.
Pitch a flat fee, not a CPM, when your numbers are small. A thousand dollars for four episodes is a proposition a small brand can evaluate; twenty-two dollars per thousand against an unfamiliar download figure is a conversation they have to learn to have.
Offer a unique discount code or a dedicated landing page. That turns your show from a brand expense into a trackable one, and a trackable channel gets renewed while an untrackable one gets cut first in a bad quarter.
And read the ad yourself, in the episode, having actually used the product. The reason host-read converts is that it is not separable from the show, and listeners can hear when it is.
What goes in a media kit that gets answered
One page. Four things.
The download figure, stated per episode, within thirty days, with the host named so the sponsor knows how it was measured. The audience: where they are, roughly how old, and what they do, drawn from your platform analytics rather than invented.
The slots you are selling and the price of each. Not a range, not on request — a number, because a range means another email before anyone can decide.
And one piece of evidence that your audience acts: a code redemption count, a survey response rate, a link click, anything that shows the show moves people rather than merely reaching them.
What does not belong in it: total lifetime downloads, subscriber counts, social media following, and five-star review screenshots. Every one of those reads as a substitute for the number the sponsor actually asked for.
Why early episodes decide the ceiling
Sponsorship rates are set against a number that compounds, and the compounding happens at the front.
A new episode's first few days decide whether the directories surface it, and whether they surface it decides how many people press play, and that is the number a sponsor will eventually buy. A show that opens quietly stays quiet, not because the content is weaker but because nothing moved it into a chart or a recommendation shelf in the window when those are decided.
That is the argument for caring about the opening days of each episode rather than the lifetime total. The lifetime total is a vanity figure that no network buys.
It is also the argument against publishing irregularly. Thirty-day windows cannot be compared across a show that posts weekly for two months and then not at all, and a sponsor planning a campaign needs to know an episode will exist on the date they paid for.
Questions people ask about podcast sponsorship
How many downloads do you need for sponsors?
Networks commonly ask for five thousand per episode in the first thirty days. Direct sponsors will go far below that if your audience is narrow enough to be worth reaching.
How much do podcast sponsors pay per episode?
At a mid-roll rate of roughly twenty-five dollars per thousand downloads, a five thousand download episode earns about a hundred and twenty-five dollars for one slot. The rate is per thousand, so the per-episode figure moves with your audience.
What is a good podcast CPM?
Eighteen to twenty-five dollars for pre-roll and twenty-five to thirty for mid-roll are the usual US bands. Host-read sits at the top of each, programmatic at the bottom.
Can you get sponsors with a small podcast?
Yes, by selling directly rather than through a network, pricing as a flat fee, and choosing advertisers whose customers match your listeners closely enough that reach elsewhere would cost them more.
Do sponsors care about subscribers or downloads?
Downloads. Subscriber counts are not comparable between platforms and cannot be audited, so no advertiser prices against them.
Is affiliate better than sponsorship when starting out?
Usually, because affiliate pays on results rather than on audience size, so it works at any scale. It also produces the conversion evidence that makes the first real sponsorship easier to sell.
Keep reading
- add plays to an episode — Sponsors buy the thirty-day window rather than the lifetime total, and you can add plays to an episode while that window is still open.
- why the feed is the only asset you own — Everything a sponsor measures runs through one file, and why the feed is the only asset you own explains how that works.
- how audio platforms actually pay — Advertising is not the only model in audio, and how audio platforms actually pay covers what the per-stream route looks like by comparison.