Most DTC founders running paid media are competing on the same three channels. Meta. TikTok. Google. The auction dynamics are punishing: CPMs on Meta have risen 30 to 40% since 2022, attribution has been fractured since iOS 14, and creative fatigue sets in within ten days of a new ad set. Meanwhile, podcast advertising, which is host-read, trust-based, and genuinely underpriced relative to the conversion quality it delivers, is being skipped by most operators.
The economics are structurally different. A host-read 60-second mid-roll on a show with 40,000 weekly listeners is not competing with 40 other brands in a real-time auction. The host reads your copy in their own voice, with their own credibility, to an audience that has listened to them for years. Edison Research's 2025 Infinite Dial report found that 68% of podcast listeners take action after hearing an ad from a trusted host. The typical click-through rate on cold Meta video is 1 to 3%.
I have tested podcast advertising across six DTC clients in drinks, functional wellness, and beauty. The first-purchase CAC sits 40 to 60% above Meta, and that comparison always makes the channel look expensive at first glance. But the 12-month LTV for podcast-acquired customers is consistently 30 to 45% higher than customers acquired through cold paid social. The host pre-sells the brand story before the customer ever sees your site. This guide covers exactly how to select shows, structure deals, brief hosts, and measure results.
Why host-read ads work differently from every other paid channel
The fundamental dynamic is trust transfer. A podcaster who has built an audience over three or five years has accumulated something no amount of paid media budget can replicate: a genuine relationship with their listeners. When they read a 60-second ad in their own voice, with their own inflection, and often their own personal experience of the product, that trust transfers partially to the brand. This is not a metaphor. IAB data shows host-read placements generate three to four times higher ad recall than pre-produced spots bought programmatically on the same platform.
This is mechanically different from display, social, or pre-roll video. Display advertising interrupts. Social advertising competes with the feed for attention it never fully captures. Pre-roll gets skipped within five seconds. Host-read podcast ads are listened to by the majority of audiences because skipping requires active effort and most listeners trust that a host who values their relationship with the audience will not endorse rubbish. The category fit is also significant. Podcast listeners skew educated, professional, and higher income than average social media users. For brands in functional wellness, premium beverages, beauty, or any product with an explanation-heavy story, this is the right demographic.
The caveat is that podcast advertising works badly for products that require visual demonstration, impulse-purchase items, or anything where the buying decision is purely aesthetic. A premium supplement, a functional drink, a skincare product with a specific mechanism, or a brand with a strong founder origin story are all well suited. A fast fashion brand or a purely visual home decor product is not.
The five criteria for choosing the right shows
Show selection is where most first-time podcast advertisers go wrong. Download counts are visible and feel like a safe proxy for value. They are not. A show with 80,000 downloads of a broad true crime audience will perform worse for a functional wellness brand than a show with 12,000 downloads of a highly engaged running and performance audience. These are the five criteria that actually predict performance.
Psychographic audience match, not demographic match
Demographics tell you who listens. Psychographics tell you why they would buy your product. A show about personal finance demographics may look ideal, but if the audience is focused on frugality rather than health optimisation, a premium supplement has no natural place in their lives. A smaller show about endurance sports, biohacking, or clean eating will have a psychographic audience that is already bought into the value of spending on functional products.
Before reaching out to any show, listen to three to five episodes. Read the top 50 reviews on Apple Podcasts. Get a feel for who the listener is and what they believe. The question to answer is not 'could this audience afford my product?' but 'does this audience already believe in what my product is about?'
Downloads per episode, not total subscribers
Subscriber counts on podcast platforms are largely meaningless. A subscriber may not have listened to an episode in two years. The metric that matters is downloads per episode over the most recent 30 to 90 days, and specifically the average across the last 10 to 20 episodes rather than a single viral outlier.
Any reputable show willing to do business will share this number. If a host refuses to provide download data, walk away. Established shows with 15,000 to 50,000 downloads per episode sit in the sweet spot for most DTC test budgets: large enough to generate meaningful conversion data, small enough that the CPM is negotiable and the host relationship is personal rather than managed by an ad sales team.
Ad load and sponsor quality
Count how many sponsors appear in a typical episode. Two to three sponsors per episode is healthy. Five or six is a sign the host has commoditised their ad slots, and listener tolerance for ad skipping is much higher in those shows. Also look at who else is advertising. If every sponsor is a mattress brand or a generic VPN, your functional wellness brand will stand out but the ad fatigue context is not ideal.
Shows where the host visibly vets their partners, mentions personal use, and limits ad load to one or two slots per episode command higher CPMs for good reason. These shows convert at meaningfully higher rates and the data supports paying the premium.
Host engagement with sponsors
The best podcast advertising results come from hosts who are genuinely interested in what they advertise. Before committing budget, send the host a sample of the product. Ask them to use it for two weeks before the first ad reads. A host who can speak from personal experience, even briefly, converts at a higher rate than one reading a script for the first time. This is not always possible with larger shows managed by agencies, but with independent hosts running shows under 100,000 downloads per episode it is standard practice.
Scroll back through the host's social media to see how they talk about their sponsors. Hosts who mention their ad partners organically outside of ad reads are rare but extremely valuable. When you find one whose audience and product fit align, they are worth paying above market rate to secure a long-term relationship.
Rate card vs. negotiated rate
Published rate cards on podcast advertising networks are starting points, not fixed prices. Independent shows with under 75,000 downloads per episode will almost always negotiate. Performance-based deals, where you pay a lower flat rate and offer commission on tracked conversions via a unique discount code, are increasingly common and worth proposing to hosts who believe in their conversion power.
Bundle pricing across multiple episodes is also standard. A six-episode package almost always comes in at a lower effective CPM than buying single episodes at the rate card. If you're testing a new show, start with a four to six episode commitment at a bundled rate. This gives you enough data to assess the channel properly without locking into a long-term contract before you know what the numbers look like.
The three deal structures and when to use each
Podcast ad deals come in three main structures. Each has a different risk profile and is appropriate at a different stage of your relationship with a show.
CPM / rate card
For first tests with larger showsStandard rate card pricing is calculated per thousand downloads. Mid-roll 60-second placements typically command £25 to £45 CPM, with pre-roll at the lower end. This structure gives you clear cost visibility before you commit, and it is the easiest to compare across shows of different sizes. The risk is paying premium rates for an audience that has not yet been validated against your specific product. Use CPM pricing for initial tests on shows with 50,000+ downloads per episode, where the network or show management runs the negotiation. Expect limited flexibility on the rate.
Flat rate per episode
Best for mid-size independent showsFor shows with 10,000 to 50,000 downloads per episode, flat rate per episode is the most common deal structure and the most negotiable. The rate is fixed regardless of how many downloads the episode receives. This works in your favour if the show has a loyal and active audience, because you benefit from backlist downloads, social shares, and episode clips. Flat rates for a 60-second mid-roll on a 25,000-download show typically range from £500 to £1,000. Negotiate a six-episode bundle at a 15 to 20% discount from the per-episode rate as a starting position.
Performance / affiliate structure
For shows with proven audiences and willing hostsPerformance deals pay the host a lower base fee plus a commission on tracked conversions via a unique discount code. For a DTC brand with good margins, this is an excellent structure because your cost is directly tied to revenue generated. Some hosts resist this model because it introduces income uncertainty. Others welcome it because they believe in their conversion power and want to share the upside. When a host has a genuinely engaged audience and personal belief in the product, propose a hybrid: a modest flat fee per episode plus 10 to 15% commission on tracked conversions. This aligns incentives and tends to produce the most motivated ad reads.
Briefing the host: what to give them and what to leave out
The brief is the most important creative asset in a podcast campaign. A badly briefed host produces an ad that sounds like a script. A well-briefed host produces an endorsement. There is a significant conversion difference between the two.
Provide the following: the three to five most important things about the product (the mechanism, the proof, the differentiation), the specific offer and discount code, the URL to send people to, and the call to action. Then stop. Do not write a script. Do not write sample copy that the host reads verbatim. The moment the ad sounds scripted, the trust dynamic breaks and it performs like produced audio rather than a host endorsement.
The elements that should be consistent across every read: the product name, the discount code (so you can track it), the vanity URL, and the core claim. Everything else, the tone, the personal story, the comparison to other products, the anecdote, should be the host's own language. If you send a product sample before the first episode, the host often includes a genuine personal experience. That is worth more than any scripted copy you could write.
After the first episode, listen back. Take note of what the host chose to emphasise and what resonated with the audience in the episode comments or reviews. Feed that information back to them before episode two, not as a correction but as data: "we noticed listeners responded well to the story about X, it would be great if you hit that again." Good hosts appreciate the feedback and it improves subsequent reads.
Attribution: how to know if it is working
Podcast attribution is not precise. Accept this upfront. It is far more robust than most brands expect, but it will never give you the clean last-click data you get from a Google Shopping conversion. The goal is a directionally accurate read rather than an exact number, and there are four data sources that give you that.
The primary attribution tool is a unique discount code for each show and each placement. When a listener uses that code at checkout, the conversion is attributed to that campaign with high confidence. The limitation is that some customers will use the code and others will not, either because they forgot it or because they found the site through a search after hearing the ad. Expect discount code tracking to capture 40 to 60% of actual podcast-attributed conversions. It undercounts, but it undercounts consistently, so the comparative data across shows is still meaningful.
The second data source is the post-purchase survey. Ask every new customer "how did you hear about us?" with podcast as an explicit option. The combination of discount code data and survey data gives you a coverage figure that is typically 65 to 80% of actual conversions. The third source is direct and branded search volume. Podcast advertising drives search behaviour. A campaign generating strong results will show a measurable increase in branded searches during the episode release window, visible in Google Search Console. The fourth source is comparing new customer acquisition rate during campaign periods against baseline, controlling for other media activity.
Use all four sources together rather than relying on any single one. Build a simple tracker with episode air dates, download counts, discount code redemptions, survey responses, and branded search volume. Review it weekly during the campaign. The pattern of where conversions cluster across shows is what tells you which to continue and which to cut.
The six-week test protocol
A single podcast episode gives you almost no useful data. Listeners catch up at different times, word of mouth from the episode takes time to propagate, and the repeat-exposure dynamic means many first-time listeners act on the third or fourth mention of a brand. The minimum viable test is six weeks across two to three shows simultaneously.
Identify, outreach, and brief
Research shows using the five criteria above. Aim for three to four candidates. Reach out directly to hosts or via their hosting platform contact, send a brief description of the brand and what you're looking for, and request download data. Send product samples to any host who expresses interest. Negotiate deal structure and lock in start dates. Prepare unique discount codes and a dedicated landing page for each show if your budget allows it.
Live episodes and weekly tracking
Each show should air one to two episodes per week during the test period, giving you four to eight ad placements per show. Track discount code redemptions weekly. Monitor branded search volume in Search Console. Run post-purchase surveys to every new customer. Do not intervene mid-campaign unless a host has made a factual error. The brief feedback loop happens after the first episode on each show, not during the campaign.
Evaluation and decision
Compile all four data sources per show. Calculate attributed CAC from discount codes as a floor, knowing actual CAC is lower due to non-code conversions. Calculate estimated actual CAC by applying the 50-60% code capture rate as a correction. Compare LTV proxies by looking at which podcast customers have returned to purchase a second time within 30 days. Any show with an attributed CAC below your payback-period target at 12-month LTV is worth continuing. Shows that are well above it should be paused and replaced with new candidates.
When to scale and what scaling looks like
Podcast advertising does not scale the way Meta does. You cannot double spend on a show and get double results. The audience size is fixed. Scaling means adding more shows, not increasing spend on the same ones. Once you have two or three validated shows, the playbook is to find five more in adjacent audience niches and run the six-week test protocol again.
There is a point where building and managing a portfolio of individual show relationships becomes unwieldy. At 10 to 15 active show relationships, most brands bring in a podcast ad agency or buying platform to manage the operational overhead. Platforms such as Acast, Spotify Audience Network, and Podscribe allow programmatic buying across many shows simultaneously, with some host-read inventory available at scale. The CPMs are higher and the relationship dynamics are different, but the operational lift is lower.
The long-term signal that podcast advertising is working is the branded search trend. Brands that run podcast advertising consistently over six to twelve months see a measurable uplift in branded search volume. This has a compounding effect on efficiency across all other paid channels, because some of the demand that podcasts create converts through paid search, direct, and email rather than through the podcast code. The channel builds brand recognition in a way that click-based channels do not, and that recognition lowers CAC across the entire acquisition mix over time.
Know which acquisition channels are working for your brand
The free scorecard takes three minutes and covers your paid media mix alongside email, conversion rate, and offer structure. It will show you where your biggest constraint is right now and whether adding a new channel like podcasts makes sense for your stage.
If you want a proper review of your acquisition strategy, the Brand Growth Audit covers your complete media mix with a prioritised action plan. Three days, Loom walkthrough, written report. No obligation to continue.
Frequently asked questions
How much does podcast advertising cost for a DTC brand?
Host-read podcast ad CPMs typically range from £18 to £45 per thousand downloads, depending on show size, niche, and ad placement. Mid-roll 60-second placements command a premium over pre-roll. For a show with 30,000 downloads per episode, a single mid-roll placement costs approximately £540 to £1,350. Many shows also offer flat-rate packages for a series of episodes, which can reduce the effective CPM for brands committing to a longer run.
How do you track conversions from podcast advertising?
The primary attribution method is a unique discount code or vanity URL for each show and placement. When a listener uses that code at checkout, the conversion is directly attributed to that campaign. You should also track post-purchase survey responses asking 'how did you hear about us?' and monitor direct and branded search traffic during and after a campaign. Podcast attribution is inherently imprecise, but discount code tracking combined with survey data gives you a reliable enough read to make optimisation decisions within four to six weeks.
What types of DTC products work best for podcast advertising?
Products that benefit from a 60-second explanation perform best on podcasts. Functional supplements, premium beverages, skincare with a specific mechanism, and any brand with a strong founder story convert well because the host can communicate nuance that display ads cannot. High-ticket or complex products that struggle to convert on cold social often find podcast advertising more efficient. Impulse-purchase or purely visual products, such as fashion or home decor, typically see lower returns.
Should you use host-read ads or produced spots for DTC podcast campaigns?
Host-read ads consistently outperform produced spots on recall and conversion for DTC brands. IAB data shows host-read placements generate three to four times higher recall than pre-produced audio. The reason is trust transfer: the host is endorsing the product in their own voice to an audience that trusts them. Produced spots can be useful for programmatic podcast buys at scale, but for direct show sponsorships, host-read is the default. The brief you give the host matters enormously, the key points to hit, the code, the call to action, and then leave the delivery to them.
How many podcast episodes should you buy to test a channel?
A minimum of four to six episodes on the same show is the standard test threshold for podcast advertising. Podcast listener behaviour means some subscribers listen weeks behind the release schedule, and the brand story builds with repeated exposure. A single episode rarely gives enough data. For a robust test, commit to six episodes across two or three shows simultaneously, track discount code redemptions weekly, and evaluate at the end of the six-week window before deciding whether to continue, adjust the brief, or redirect budget.
What is a reasonable podcast advertising CAC for a DTC brand?
Podcast-attributed CAC for DTC brands typically sits 40 to 60% above the Meta blended CAC for the same brand. On a straight first-purchase basis this makes podcast advertising appear expensive. The case for the channel is in LTV: customers acquired through host-read podcast ads consistently show 30 to 45% higher 12-month LTV because they arrive pre-sold on the brand story and have higher intent. If your brand has strong repeat purchase rates and you know your LTV with confidence, podcast CAC is often justified on a payback-period basis within six to nine months.
About the author
Caner Veli founded and exited Liquiproof, scaling from zero to 3,000+ retailers globally in under 6 years. He now runs Purposeful Profits, a focused growth consultancy for founder-led DTC and CPG brands. 12 named sprint clients. 518% average growth. 27x highest ROAS. Read more about Caner →