How to Start a Podcast: The Complete 2026 Blueprint.

Podcasting in the United States is a large listening market attached to a small and highly concentrated revenue market. Edison Research's Infinite Dial and Podcast Consumer work puts monthly podcast listening at roughly 47% of Americans age 12 and older, with about one third listening weekly. Global podcast revenue was estimated at around $7.3 billion in 2024, and one broader market sizing put the total podcasting market near $31.29 billion in 2024. Those are ecosystem numbers. They include Spotify, iHeart, Wondery, NPR, and the top few hundred shows. The median independent show captures effectively none of it in year one.

There are three buyers who actually send money to a small US podcast. First, advertisers and sponsors, who buy on downloads at host-read CPMs commonly quoted in the $18 to $50+ per thousand downloads range, though show-level 2026 CPM data is not published in any primary source and should be treated as uncertain. Second, the audience directly, through Patreon-style memberships, courses, live events, and merchandise, which is where most shows under 10,000 downloads per episode make real money. Third, and most reliably, a business that funds the show out of its own marketing budget because the podcast generates qualified leads, in which case the show is a cost center measured on pipeline, not a P and L of its own.

The practical consequence for someone spending money next month: pick your buyer before you buy a microphone. A show built for CPM sponsorship needs scale and needs 6 to 18 months to reach the download volume a network or direct sponsor will underwrite. A show built for a B2B services firm can be profitable at 300 listeners if those listeners are the 300 procurement managers who buy your $80,000 engagement. A show built for audience-direct revenue needs a niche narrow enough that 1,000 people will pay $5 to $10 a month. These three businesses share equipment and share almost nothing else.

Trend to price in for 2026: video. YouTube and Spotify video distribution has moved from optional to expected in most commercial categories, which adds camera, lighting, and video editing cost to a budget that used to be audio-only. That single shift is the main reason the professional tier has drifted from roughly $500 toward the $5,000 end of the range, and it is the most common reason a founder's original budget turns out to be half of what gets spent.

The physical siting question for a podcast is not foot traffic, it is reverberation time and noise floor. A spare bedroom with a closet, carpet, and full bookshelves outperforms a large open office with hard walls, and it outperforms a rented studio you can only reach twice a month. Budget $50 to $300 for blankets, rugs, and DIY absorption, or $500 to $2,000 for broadband panels, bass traps, and door seals if the room has a slap echo you cannot kill. The real constraints are HVAC noise, refrigerator hum, road traffic, and neighbors. Test the room by recording 60 seconds of silence and looking at the noise floor; anything above roughly -60 dBFS will be audible under your voice on cheap earbuds.

A dedicated studio build-out at $1,500 to $13,000 makes sense in exactly two cases: you record multi-guest video regularly, or you plan to rent the room to other shows. Renting a commercial studio at hourly rates works for a limited-run series but breaks a weekly schedule, because the scheduling friction is what causes missed weeks and missed weeks are what kill shows. Remote recording through Riverside, Zencastr, or SquadCast records each participant locally and sidesteps siting entirely, at the cost of guest technical failures that you will have to troubleshoot live.

The channel decision is now the bigger one. Audio-only RSS distribution through your host to Apple Podcasts, Spotify, Amazon Music, and iHeart is table stakes and costs $0 to $50 per month. YouTube has become a primary discovery surface for interview and commentary shows, which means a camera, lighting, and a video edit added to every episode. That decision roughly doubles per-episode production time and pushes outsourced editing toward the top of the $50 to $200 per episode band. If your buyer is a B2B client, video is close to mandatory because clips are what circulate on LinkedIn. If your buyer is a commuting audio listener in a narrative or news niche, video may be pure cost with no return.

Price the unit as one episode. A self-produced audio episode has near-zero marginal cost: hosting is a flat $0 to $50 per month regardless of downloads, so the direct cash cost per episode is a few dollars. Outsource the edit and it becomes $50 to $200 per episode, or roughly $200 to $800 per month at weekly cadence. Add show notes, transcript, and social clips and outsourced packages run in the range of $326 to $1,181 per month for four episodes. Agency-produced business podcasts with video, booking, and strategy run $2,000 to $20,000+ per month. The cost you will not put in the spreadsheet is your own: 6 to 10 hours per episode, which at any honest hourly rate exceeds the cash cost by a wide margin.

Now the revenue side. Host-read sponsorship is commonly quoted at $18 to $50+ per thousand downloads, though show-level 2026 CPM data is not published in any primary source and this range should be treated as uncertain. At a 25 dollar CPM, a show doing 1,000 downloads per episode with one mid-roll earns about $25 per episode, or roughly $100 per month. That does not cover a $200 to $800 outsourced edit. To cover a fully outsourced weekly show on ad revenue alone at that CPM, you need somewhere in the range of 8,000 to 30,000 downloads per episode, which is why so few independent shows are self-funding.

Because direct episode costs are small relative to revenue once scale arrives, gross margin on a creator-owned show is structurally high, plausibly above 60% to 70%, though this is an inference from reported cost structures rather than a published podcast-specific figure. Net margin is the volatile number. A solo creator with lean costs can be net positive at modest scale. An agency-style operation with staff, a studio lease, and booking overhead behaves like any creative services firm, where efficiently run businesses tend to land in the 10% to 25% net range. No SBA or BLS category tracks podcast business margins, so treat every margin number here as directional.

The economics change completely if the buyer is your own company. A B2B show costing $1,500 per month all-in that sources two qualified opportunities per quarter at an $80,000 average contract value is not a media business at all; it is marketing spend with a payback measured in one closed deal per year. Founders who make money in podcasting overwhelmingly get there through this route or through direct audience monetization, not through CPM.

At launch a podcast has no employees. The founder is host, producer, and booker, and the only outside role that pays for itself immediately is an audio editor working per episode at $50 to $200. Hire that person before you hire anyone else, because editing is the task most likely to make you skip a week, and skipped weeks are the leading indicator of a dead show. Vet by sending a raw 20 minute file with a cough, a crosstalk section, and a bad guest connection and comparing what comes back. Editors who use iZotope RX for noise repair and who deliver to a consistent -16 LUFS for stereo or -19 LUFS for mono loudness target are worth the top of the range.

The second hire is a producer or booking coordinator, typically part-time at $20 to $40 per hour or on a per-episode retainer, once you are booking two or more guests a month. Their job is outreach, calendar management, pre-interview briefs, and the release paperwork. The third, if you commit to video, is a video editor or clips editor, often a separate specialist from your audio editor, priced per episode or per clip package. Only at real scale does a show justify a full-time producer, and at that point you are running a small media company with the payroll, workers' compensation, and unemployment insurance obligations that follow.

Two roles founders wrongly try to hire early. A sales rep for sponsorships is useless below the download volume sponsors will buy, and a podcast ad network or agency will not take you on until you are consistently in the thousands of downloads per episode. A social media manager cannot manufacture an audience for a show with no distinctive point of view. Both hires are more commonly the symptom of avoiding the real work, which is publishing consistently and booking better guests. Keep the paperwork tight regardless of headcount: guest releases granting perpetual rights to the recording, work-for-hire or assignment language in every editor contract, and correct 1099-NEC filing for contractors paid $600 or more in a calendar year.

Days 1 to 30, the goal is a functioning feed and a listener base you can name. Publish three episodes at once so a new listener has something to binge, then personally message every relevant person in your existing network with a specific ask: listen to episode two and tell me if the third question was worth asking. Guests are your distribution. Each guest should receive a ready-to-post package the day the episode drops: a 60 second vertical clip, three pull quotes, the artwork, and the link. Guests who post reach audiences you cannot buy. Guests who are never asked, never post. Track which guest posts drove downloads; that is your first real acquisition data.

Days 31 to 60, go where the audience already gathers. For B2B shows that is LinkedIn and industry newsletters; for consumer niches it is the relevant subreddit, Discord, or Facebook group, where you participate as a member and only occasionally link. Cross-promotion swaps with shows of similar size are the highest-return free channel in podcasting: trade a 30 second promo read in each other's mid-roll. Pitch five shows in your topic band. Paid promotion at this stage, $0 to $500 per month in social ads or audiogram tools, generally buys plays rather than subscribers and is worth testing only with a tracked link.

Days 61 to 90, convert. If your buyer is a B2B client, the podcast itself is the sales motion: every guest is a prospect or a referral source, and a warm follow-up two weeks after their episode airs is where deals begin. If your buyer is the audience, launch a $5 to $10 per month membership with one concrete benefit, an ad-free feed or a bonus episode, and expect a 1% to 3% conversion of engaged listeners at best. If your buyer is a sponsor, do not wait for a network. Identify 20 companies already advertising on shows in your category, email the marketing manager directly with your IAB-certified download numbers, audience description, and a flat-fee offer for a four-episode package. Flat fees are easier to sell than CPM at small scale, and a signed insertion order with makegood terms is the document that turns a hobby into a business.