The honest guide to podcast sponsorship

Start with the number everyone quotes

Sponsorship deals are priced on downloads, so it's worth knowing exactly what a download is — because the word hides a lot. Under the IAB's Podcast Measurement Technical Guidelines (v2.2 is current), a download only counts when the audio is actually pulled for at least one minute of content, measured by byte range. Bots and known crawlers are filtered out, partial and malformed requests are dropped, and repeat requests from the same listener inside a 24-hour window collapse into one. Play an episode twenty times before lunch and you're still a single download; come back tomorrow and it's a new one.

Two words matter here. IAB-compliant means a host measures this way. IAB-certified means an independent auditor has formally verified that they do — a paid, ongoing process, and the stronger signal. When someone quotes you an audience size, ask which they mean. "Downloads" and "listens" are not the same thing, and no honest host will promise that every counted download reached a pair of human ears.

The positions, and what they're worth

An episode has three ad slots, and they are not interchangeable:

• Pre-roll runs in the first minute or two. It guarantees exposure — everyone who presses play hears it — but attention is still warming up. Typical range: $18 to $25 CPM.
• Mid-roll lands where listeners are most engaged, which is why it's both the most expensive and the most effective. Typical range: $25 to $40 CPM, higher for premium host-reads.
• Post-roll runs at the end. It's the cheapest — often 40 to 60% below mid-roll, in the $10 to $20 range — because it reaches only the most dedicated listeners, but those are frequently your most valuable ones.

There's no universally "best" slot. Mid-roll wins on raw conversion; post-roll can win on cost-per-acquisition for a loyal niche; pre-roll wins when guaranteed reach matters more than depth.

Host-read versus programmatic: the real fork

This is the decision that most changes what you pay and what you get. A host-read is the host personally vouching, in their own voice and cadence. A programmatic or dynamically inserted ad is a produced spot stitched into the slot by a server — often a different creative for different listeners based on geography, device, or app — with frequency caps to avoid fatigue.

The performance gap is well documented. Host-reads command $25 to $50 CPM against programmatic's $5 to $15 because they convert three to five times better on the same show, and Podscribe's Q2 2025 benchmark found host-reads beat producer-read spots by 31% on purchase rate. The tradeoff is honest: programmatic scales effortlessly and targets precisely; host-reads don't scale but carry trust that no server can insert.

You are not buying a slot. With a host-read you're renting a relationship — which means the fastest way to waste the money is to make the host say something they don't believe.

Baked-in versus dynamic: who hears it, and forever

One more mechanical fork. A baked-in ad is edited permanently into the audio file — every listener, forever, including someone who finds the episode in three years, hears the same spot. Great for evergreen sponsors, a liability when the offer expires. A dynamically inserted ad lives in a marked slot and is stitched in at download time, so it can be swapped, targeted, capped, and retired. Most major publishers now run host-reads through dynamic insertion to get both the trusted voice and the flexibility. Ask which you're getting; it determines whether your "limited-time" offer is still playing long after it ended.

Measure it honestly, or you'll fool yourself

Podcast attribution is genuinely hard, and the easy tools flatter you in the wrong direction. Promo codes are cheap and require no technology, but they capture only about 21% of real conversions — most people who heard your ad buy later, on another device, and never type the code. Lean on codes alone and you'll cancel winning campaigns for looking weak.

The honest stack combines methods:

• A vanity URL or unique code for the direct, high-intent signal.
• A tracking pixel on your site to match visitors back to ad exposure. Podscribe's Q2 2025 data found pixel attribution surfaced nearly 7x more conversions than surveys and over 4x more than promo codes.
• A brand lift study for awareness goals with no immediate click — Nielsen puts average podcast lift at +10 points on awareness, +8 on information-seeking, and +6 on purchase intent.

Use at least two. One number in isolation is a story you're telling yourself.

Stay on the right side of the law — and the listener

If money changed hands and it isn't obvious, disclose it. The FTC's standard is clear and conspicuous, and because podcasts are consumed by ear — often while driving or doing dishes — the disclosure has to be spoken. A line buried in the show notes doesn't count. The penalty reaches up to $53,088 per violation in 2025, but the bigger cost is trust: audiences reward hosts who are upfront and abandon ones who feel slippery. A three-word cue at the top of the read handles the whole obligation.

How matching should actually work

The last honest point is about fit. A sponsorship works when the product genuinely belongs in front of that audience — and forcing a mismatch burns the host's credibility and your budget together. That's the logic behind 10AM Media's Ads Marketplace: it matches advertisers and shows by meaning, not just category tags, and creators keep final say over what runs under their voice. Pair a well-matched show with a host-style script written for the ear — which Ad Studio can draft, voice, and design from a single brief — and honest measurement behind it, and you have the parts of a sponsorship that actually earns its premium instead of just charging one.