# Cannabis Podcast Marketing: The Channel That Is Not Throttled **Author:** John Morabito (Founder, /winston) **Published:** September 19, 2026 **Reading time:** 12 minutes **Canonical:** https://www.winstondigitalmarketing.com/playbooks/cannabis-podcast-marketing/ Every cannabis marketing conversation reaches the same wall eventually. The major ad platforms restrict the category and enforce their rules unevenly, social accounts get limited or removed with an appeals process that goes nowhere anyone can name, and the money that would be paid media in any other industry sits in the budget with fewer places to go than the plan assumed. Somebody says "what about a podcast." The room agrees it is a good idea. Nothing happens for eighteen months. The idea deserves better than that. What makes it work is structural, and it has very little to do with how good your content is. Podcasting still runs on RSS. The feed is a file on a server you control, and the directories that list it point at your file rather than hosting your content. No algorithm decides whether today's episode reaches your subscribers, because there is no algorithm in the path. Nothing reviews an episode before it publishes. In a category that has spent a decade being removed from other people's platforms, the distribution layer of podcasting is one of the last pieces of consumer media where nobody in the middle has the power to cut the line between you and the people who asked to hear from you. That is the argument, and it is worth being exact about where it stops, because the part of podcasting that takes your money is roughly as restricted as everything else. > **General information, not legal advice** > > This playbook is about channel strategy. It is not legal advice. Cannabis advertising and promotion rules are set state by state, they change often, and a branded show or a paid sponsorship can count as advertising under them. Before you launch a show or sign a sponsorship, have your own counsel review the plan and check your state regulator's current guidance. Nothing here should be read as a statement about health effects. ## Three layers, and only two of them are open "Podcasting is open to cannabis" is true and too coarse to plan against. Split the medium into three layers and the picture gets usable. - **Distribution is open.** RSS is a published format and a feed is a file. Apple, Spotify, Pocket Casts, Overcast and the rest read that file on a schedule. There is no moderation queue between recording an episode and it landing in a subscriber's app, and no engagement threshold to clear. - **Publishing is mostly open, with conditions.** Getting listed in a directory is a submission, not a right, and the major directories apply their own content guidelines covering explicit ratings, the accuracy of your show metadata, and material aimed at or likely to appeal to minors. Shows about the industry, the regulation, the science, and the culture are published widely. A show built around depicting consumption is on much thinner ice. - **Buying is restricted.** The advertising marketplaces and representation networks that sell inventory across thousands of shows keep their own lists of advertiser categories they decline, and cannabis appears on plenty of them. The self-serve path is largely closed the same way it is elsewhere. So the honest version of the pitch is narrower and still good: the layers you cannot buy your way into are the ones that stayed open. Earned attention on other people's shows, and owned attention on your own. That is the whole of this playbook. Worth saying plainly, because the optimism in this category tends to outrun the facts: a podcast is not a workaround for a platform policy. If a network declines the category, the answer is a different network or a direct deal with an individual show, not a creative reading of the policy. And a show does not exempt you from your state's advertising rules. Several states set requirements about the audience a cannabis ad may reach, and a sponsorship on a national show is exactly the kind of buy those rules were written about. ## Play one: get on other people's shows Guesting is the cheaper play and for most operators it should come first, because it tests whether you have anything to say before you commit to saying it every two weeks forever. You borrow an audience someone else spent years building, you produce nothing, and you find out quickly whether the thing you believe about your category is interesting to anyone outside your company. ### Which shows actually reach your buyer Most guesting programs waste their effort on the wrong tier, which usually means chasing the largest consumer show in the category and getting no reply. Four buckets are worth working, and which ones depend entirely on who buys from you. - **Industry and B2B shows.** If you sell into dispensaries, run an ancillary business, or supply the trade, this is the whole game. The audience is small, the listeners are the buyers, and the hosts need episodes constantly. - **Regional and market shows.** State-level cannabis coverage tracks licensing, regulation, and the local operators. For a retailer or a single-state brand, one episode here reaches more actual customers than a national show with fifty times the downloads. - **Consumer and culture shows.** Harder to book, useful for brand, and the least measurable. Worth pitching once you have a track record, worth ignoring while you do not. - **Adjacent shows outside the category.** Small business, retail operations, supply chain, food and beverage, local media. These are underworked, and you arrive as the only guest who can speak plainly about operating in the category. The hosts are usually more interested than the cannabis shows are, because it is not their daily subject. Find them the way a listener would. Search the directories for your state name and the category, look at what the guests you respect appeared on recently, and check the shows your own customers mention. A show that published last month and has more than a handful of episodes is bookable; a show whose last episode is from 2024 is a directory listing with no audience behind it. ### What makes a pitch bookable A host books a guest because that guest solves the problem of what to publish in two weeks. Every rule below follows from that. The pitch has to contain the episode. - **One specific argument, stated in the first line.** Something you can hold for forty minutes and that most guests in your category would not say. "Most dispensary menus are invisible to search engines and here is how to check yours in a minute" is an episode. "Discussing the future of cannabis retail" is not. - **The reason you in particular get to say it.** One line. What you run, how long, at what scale. Not a biography. - **Evidence you listened.** Name an episode and say what you would add to it. This takes fifteen minutes and separates you from everything else in the inbox. - **Three questions the host can use verbatim.** You are writing part of their prep. They will notice. - **Nothing attached.** No deck, no press release, no media kit, no one-sheet with your logo at the top. The pitches that fail are consistent. The launch-timed pitch fails because a product launch is your news and not the host's episode. The founder-story pitch fails because the story is interesting to the founder. The mass send fails because it is visible as one from the first sentence. And the pitch that leads with audience size fails because the host already knows how big their show is. ### The part everyone skips An appearance is worth roughly double what most brands get out of it, because most brands treat publication as the end. When the episode drops, ask the host for the transcript, publish a page on your own site about the conversation that links to the episode on their domain, pull two or three segments for your owned channels, and send it to your email list as something you made. The audio stays on their site. Put the written record of what you said on yours. ## Play two: run your own show Running your own is a materially bigger commitment than the planning meeting suggests, and it is worth being blunt about the cost before the format conversation, because the format is mostly a decision about cost. ### What an episode actually takes The recording is the part everyone pictures and the smallest part of the work. A published episode is: booking and confirming a guest, prep and a question outline, the recording itself, an edit, artwork and metadata, show notes with links, a transcript cleaned up from a machine draft, an episode page on your site, and the promotion afterward. Nearly all of that lands on one person, and if that person's job description does not have it in writing, it competes with their actual job and loses by the third month. The format decides most of that bill. A solo or co-hosted show removes guest booking, which is the single most unpredictable line. An interview show buys you other people's audiences and other people's calendars at the same time. A roundtable with the same three internal voices is the cheapest to sustain and the hardest to make anyone care about. Pick on the basis of what you can staff for two years. ### Why most brand podcasts die They die in a small number of recognizable ways, and every one of them is a decision made before the first episode. - **The cadence was chosen from ambition.** Weekly is picked in the enthusiastic meeting, then meets a recall, a compliance deadline, and a staffing gap in the same quarter. Pick the cadence you can hold in a bad month. - **Nobody owns it.** An unowned show is everyone's fourth priority. It needs a name in a job description and time on a calendar. - **It got measured as an acquisition channel in month three.** Podcasts do not produce attributable last-click revenue on that timescale, and they are usually not supposed to. A show judged on that will be killed correctly by its own metrics and wrongly by its own framing. - **It is a commercial with intro music.** Nobody subscribes to a commercial. If every episode arrives at your product, listeners work that out around episode four and the show becomes a feed nobody opens. - **The guest pipeline ran dry.** Four enthusiastic friends, then silence. Build the list out to twenty before you publish episode one. The inverse is the test worth applying before you start. Would this show be worth making if it produced zero attributable sales? If the answer is no, do not start it, because for the first year the answer will be no in the data whether or not it is true in reality. If the answer is yes, because it documents a category nobody is documenting well, or it gets your operators talking to the people they buy from and sell to, then it will survive long enough for the compounding part to start. ## The compounding part: episodes become text A podcast earns its place in a search and visibility program for one mechanical reason. Audio is invisible to the systems that decide what gets found. Crawlers do not listen, and AI retrieval systems work over documents. An episode that exists only as an audio file in a feed contributes nothing to either. The conversion step is text, and it is not optional. Every episode gets a page on your own domain carrying a cleaned full transcript with speaker labels, a written summary that answers the question the episode is about, show notes with real links, and timestamps for the sections. Clean the machine transcript before it goes up. The errors cluster on exactly the terms you care about, which in this category means product names, cultivars, regulatory bodies, and license types. One hour of two informed people talking produces several thousand words of specific, quotable, first-hand text, covering the questions people actually ask in the order they actually ask them, in a category where most of the published writing is thin and repetitive. It is the cheapest genuinely original long-form content a cannabis operator can produce, and the shape it comes in happens to be the shape retrieval systems prefer. The formatting decisions that determine whether those pages get quoted are a subject of their own, covered in [podcast SEO for AI citations](https://www.winstondigitalmarketing.com/playbooks/podcast-seo-for-ai-citations/). The second compounding effect is the host. A named person who turns up repeatedly across shows in a subject, with a transcript record attached to their name each time, becomes something search engines and assistants can associate with that subject. Brands are hard to establish as experts, and people are comparatively easy. This is the cheapest expert presence available to a cannabis business, and it is why the guesting play keeps paying after the download numbers are forgotten. ## What a cannabis show cannot say Speech is not a lighter regulatory category than copy, and a transcript is a document. The rules you follow on a product page follow you into a recording. - **Health claims are health claims out loud.** The federal position on marketing a product as treating, curing, preventing, or mitigating a condition does not change because the words were spoken. Once you publish a transcript, they are written down too. - **Guests say things.** This is the risk that is specific to the medium. A guest with an opinion about what their product does for a condition has handed you a claim under your brand. Edit it, or do not publish it. - **Sponsorships get disclosed.** A paid mention is an ad. Say so clearly in the episode and in the show notes, and hold sponsors to the same claim standard you hold yourself to. - **Your state's advertising rules still apply.** Depending on the state, that can include how the show is promoted, what it may depict, and requirements about the audience a cannabis ad may reach. Treat a branded show as advertising until your counsel tells you otherwise. - **Directory policies are separate again.** Ratings and metadata accuracy are the directories' rules, not the regulator's, and a show can be compliant with your state and still get pulled from a listing. The same reasoning that governs what a brand can post applies here, and the platform-by-platform version is in [the cannabis brand social media rules](https://www.winstondigitalmarketing.com/playbooks/cannabis-brand-social-media-rules/). ## Building it 1. Pick the play. Guesting first unless you already have a host, a producer, and twenty guests lined up. Guesting tells you in six weeks whether you have a subject. 2. Build the show list. Twenty shows across the four buckets, filtered to ones that published in the last sixty days, ordered by how close the audience is to your buyer. 3. Write one argument. The thing you can defend for forty minutes that most people in your category will not say. Everything downstream is a version of that sentence. 4. Pitch one show at a time, personally, with three usable questions and nothing attached. 5. Capture every appearance. Transcript, a page on your domain, clips for owned channels, a note to the list. 6. If you are launching your own, produce one full episode end to end before choosing a cadence, and have the person who did it report the real hours. 7. Build the episode page template before episode one: transcript, summary, timestamps, show notes, host bio. Retrofitting eighty episodes is a project nobody ever funds. 8. Have counsel review the format, the sponsorship language, and the guest release, not just the first script. 9. Measure it as reach and evidence, not last click. Appearances landed, episode pages indexed, quotes appearing in AI answers, and whether prospects reference it on sales calls. ## The point Cannabis operators spend an enormous share of their attention on the channels that keep rejecting them and very little on the ones nobody has closed. Podcasting sits in the second group because of how it was built: an open protocol, a direct subscription, and nobody in the middle with a policy team deciding whether your episode ships. The cost of entry is time, which is the resource most operators in this category have more of than permitted budget. The catch is that it pays on the timescale of owned media, which means it belongs next to the rest of the work that compounds. It sits directly on top of the program in [cannabis content marketing](https://www.winstondigitalmarketing.com/playbooks/cannabis-content-marketing/), and it feeds the same audiences that [creator partnerships](https://www.winstondigitalmarketing.com/playbooks/cannabis-influencer-marketing/) reach from the other direction. We plan and run this alongside search, owned channels, and content as part of our [cannabis marketing practice](https://www.winstondigitalmarketing.com/services/cannabis-marketing/), and if you want to know whether anything you have already published is being cited by AI assistants today, the [free AI visibility audit](https://www.winstondigitalmarketing.com/audit/) will tell you. ## Frequently asked questions ### Can cannabis brands buy podcast ads? Sometimes, and rarely through the easy route. The large podcast advertising marketplaces and representation networks maintain their own lists of advertiser categories they will not accept, and cannabis sits on many of them alongside the categories the big ad platforms also restrict. That closes the self-serve path where you pick a budget and a target audience and let a marketplace place you. What remains open is buying directly from individual shows, cannabis-specific networks, and hosts who set their own advertiser policy, which works but is a negotiation per show rather than a media buy. Policies differ by network and change, so confirm the current one with the network rather than assuming, and treat any sponsorship as advertising subject to your state's cannabis advertising rules, including audience composition requirements where your state has them. ### Can a cannabis brand publish a podcast on Apple Podcasts and Spotify? Generally yes, because listing a show is not the same as buying advertising, and the directories index a feed you host rather than hosting your content themselves. Shows about the cannabis industry, its regulation, its business, and its culture are widely published on the major directories. The directories do apply their own content guidelines, which cover things like explicit ratings, accuracy of show metadata, and content aimed at or appealing to minors, so a show that stays on the business, science, and culture side of the category is on much steadier ground than one built around depicting consumption. Because listing is not guaranteed, the practical protection is that the feed is yours: if a directory ever drops the show, the feed keeps publishing and you resubmit elsewhere, which is not true of a social account. ### How do you get booked as a guest on a podcast? By pitching an episode rather than a person. A host books a guest because that guest solves the problem of what to publish in two weeks, so the pitch has to contain the episode: one specific argument you can hold for forty minutes, a reason you in particular get to make it, evidence you have actually heard the show, and three questions the host can read verbatim. Keep it short, send it to one show at a time, and leave out the deck, the press release, and the founder origin story. The pitches that fail are the ones timed to a product launch, because a launch is your news and not the host's episode. For cannabis operators the highest-yield targets are usually industry and regional shows where your daily work is the subject, plus adjacent business shows where you are the only person in the category who will speak plainly. ### Do podcasts help SEO and AI visibility? The audio does not, and the text that comes off it does. Crawlers and AI retrieval systems read documents, so an episode that exists only as a file in a feed is invisible to both. The step that converts the show into search and citation value is publishing every episode as a page on your own domain with a cleaned full transcript, speaker labels, a written summary, and linked show notes. That produces long-form, specific, quotable text in a category where most published writing is thin, and it accumulates rather than expiring. The named host matters too, since a person who appears repeatedly across shows in a subject becomes an entity that both search engines and assistants can associate with the topic. The mechanics of formatting episode pages so they get retrieved and quoted are a separate subject in their own right. ### How often should a cannabis brand publish podcast episodes? At whatever cadence you can hold through your worst month, which for most operators is every other week rather than weekly. Cadence is the single most common cause of death for a brand podcast, because the schedule gets chosen during the enthusiastic planning meeting and then meets a product recall, a compliance deadline, and a staffing gap in the same quarter. A show that publishes twice a month for three years is a genuine asset; a show that publishes weekly for eleven episodes and stops is a page of dead links and a small amount of public evidence that the brand does not finish things. Decide the cadence after someone has actually recorded, edited, transcribed, and published one episode end to end and reported how long it took.