What does it actually cost to sponsor a small podcast?

Podcast sponsorship has a reputation as a premium channel, and at the top of the market it is. Further down, among shows with a few thousand listeners and a host who answers their own email, it is one of the few places a small advertiser can still buy attention at a price that makes sense — provided you understand what you are buying and are honest about what you will be able to measure. (Disclosure: published by BidSurvivor, which sells a competing kind of ad slot — the last section says how the two differ and where ours is worse.)

What the numbers look like at the small end

Podcast advertising is conventionally priced on CPM, cost per thousand downloads. Industry ranges have been fairly stable: roughly $15–25 CPM for a thirty-second pre-roll, $20–35 for a mid-roll, with host-read spots at the higher end and produced spots lower.

Apply that to a small show. Two thousand downloads an episode at a $25 mid-roll CPM is fifty dollars a spot. That is the whole budget for a lot of the people reading this, and it is a real, host-read placement rather than a programmatic impression.

But the CPM framing misleads at this size, in both directions.

It understates what you get. A podcast mid-roll is thirty to sixty seconds of a trusted voice with no competing ad next to it, consumed by someone who chose the show. That is not comparable to a banner impression, and pricing it as though a thousand downloads were a thousand impressions of anything else is a category error the CPM unit invites.

It overstates the audience. A download is not a listen. It counts a request for the file — which podcast apps make automatically for subscribers, whether or not the episode is ever played, and sometimes more than once. The gap is the same problem as what an ad impression actually counts, and it is arguably worse here because the counting happens in an app you cannot see.

The practical figure: expect to pay $40–150 for a single host-read spot on a show with one to five thousand downloads, and expect a wide spread, because at this size rates are set by the host's guess rather than by a market.

Many small shows have no rate card at all. Some will trade a spot for something you can do. Almost all will negotiate, because the alternative for them is an empty slot, and a sponsor who returns is worth more to them than a sponsor who paid full price once.

The attribution problem, which nobody has solved

Podcast advertising is the hardest mainstream channel to measure, and honest sellers will say so.

There is no click. The listener is in the car, on a walk, doing dishes. The path from hearing your name to visiting your site runs through their memory and a later search, which produces direct or branded-search traffic with no referrer attached to it. This is the exact case where attribution for a one-person team breaks down, and no tooling fixes it.

The standard workarounds are all partial:

Vanity URLs. A short, sayable address specific to the show. It works, and it undercounts badly, because most listeners will not type a URL from audio — they will search your brand name instead and never touch the tracked link.

Discount codes. Better, because there is a reason to remember them. They still undercount everyone who converted without using the code, and they change the offer, which contaminates the test you were trying to run.

Ask at signup. The free-text "how did you hear about us" field catches podcast mentions when nothing else does. Messy, self-reported, and frequently the only source that sees the channel at all.

Watch the totals around the drop date. Crude, and at small volumes a real spike is visible. If a show with three thousand downloads sends nothing you can detect in your weekly numbers, it sent very little, and that is a usable finding.

Assume every method you use undercounts by a wide and unknown margin. That is not a reason to avoid the channel. It is a reason not to make a large bet on it before you have run a small one.

How to test one for under $200

Pick a show by fit, not size. A niche show of eight hundred listeners in your exact category beats a general-interest show of ten thousand, and it will cost a fifth as much. Fit is the only variable at this size that reliably moves the outcome.

Write to the host directly, once, briefly. Say what you make, who it is for, why their audience specifically, and what you can spend. Small shows are used to being ignored by advertisers; a short concrete email gets replied to.

Buy two spots on the same show, not one on each of two shows. One appearance rarely registers; the second is where recognition starts. This is the same effective frequency logic that governs every other channel, and it is why single-spot tests so often report a false negative.

Give the host the facts and let them say it. Host-read means host-worded. A script read verbatim in a voice that is obviously not theirs is the version that does not work, and it is the most common way small sponsors waste the placement.

Decide the success threshold before it airs. How many signups make a second buy worth it? Write the number down first, because deciding when to stop after seeing an ambiguous result is how budgets drift for months.

Where ours differs, and where it is worse

Our board sells a different shape of the same thing: two twelve-hour slots a day on one page, at auction. The first brand into an empty slot pays nothing; a held slot can be taken from a $1.00 floor, capped at $5,000; every visitor starts with $100 of house credit before signing up.

The difference that matters for this comparison is measurement. Every slot publishes what it delivered — card opens and click-throughs, counted separately — on the front page and on each brand's own page, before anyone bids. There is a click, so there is a number, and the number is public whether it flatters the slot or not.

Where ours is worse: nobody trusts a logo the way they trust a host. A podcast sponsorship borrows credibility from a person your listener already chose to spend an hour with, and a slot on a board borrows nothing. If your product needs to be explained or vouched for, a host-read spot does something we cannot do at any price. If it needs to be found and clicked, a countable placement will tell you within a day whether it worked, which no podcast will.