Budget kills far fewer podcasts than vagueness does. Shows launch with enthusiasm, publish a handful of unfocused episodes, then quietly stop — not because the microphone was cheap, but because nobody could say who the show was for or why it existed. So before you spend a single dollar on gear, spend an afternoon on episode zero: the planning work that decides whether there is a show at all.

Episode zero: plan the show before the gear

Write one sentence that names the audience, the problem, and the format: for heads of operations at mid-size logistics firms, a 25-minute interview show about the unglamorous systems that keep freight moving. If you cannot write that sentence, you do not have a podcast yet — you have an urge. Then make three commitments. First, a cadence you can actually sustain: fortnightly and reliable beats weekly and abandoned. Second, a season model: commit to exactly ten episodes, then formally reassess, so the show has a finish line instead of an open-ended obligation. Third, a success metric that is not downloads — guests who become pipeline, prospects who mention an episode on a sales call, replies to the companion newsletter.

The cheap identity work matters too. Check that your show name is not already taken in the podcast apps and that a matching domain exists. Design artwork that survives thumbnail size — if the title is unreadable when the cover is the size of a postage stamp, it is unreadable where people actually browse.

Pick a format you can sustain, not the one you admire

Interview shows are the easiest to start: guests supply half the content and bring their own audiences when they share the episode. Their weakness is sameness — there are thousands of polite interview shows, so yours needs a sharper question set or a narrower niche. Solo shows are the cheapest and fastest to produce and build the strongest personal authority, but they demand a genuine point of view and tight scripts; fifteen strong minutes beats forty meandering ones. Narrative shows — scripted, multi-voice, documentary-style — are the most memorable and by far the most expensive, easily consuming several times the edit hours per finished minute. The pragmatic call for a first season: interview or solo. Earn the narrative season later, once the pipeline runs itself.

Recording setups by tier

Tier one is nearly free: the phone or laptop you own, recorded in a small room full of soft surfaces, plus a free remote-recording tool for guests. Acceptable for validating the idea; not where you want to live. Tier two, roughly a hundred to a hundred and fifty dollars, is the sweet spot: a USB dynamic microphone, a pop filter, and closed-back headphones. Dynamic beats condenser in untreated rooms because it rejects the echo and the neighbor's dog. Tier three — an audio interface, XLR microphone, and a treated space — is a post-episode-ten reward, not a launch requirement. Two rules outrank all gear: the room matters more than the microphone (a closet full of clothes beats an expensive mic in a glass conference room), and for remote interviews, record each side locally on the guest's own machine whenever possible instead of capturing the call audio.

An editing workflow that takes hours, not days

The trap in editing is perfectionism: hunting every um until a one-hour edit becomes a ten-hour edit nobody notices. Instead: top-and-tail the recording, cut the trainwrecks — false starts, tangents, tech hiccups — and leave the natural speech alone. Normalize loudness to about -16 LUFS so your show sits at the level listeners expect between other podcasts. Keep the music intro under fifteen seconds; nobody subscribed for your theme song. Once you have a template — intro, music, outro, export settings saved — a competent interview edit settles at roughly one and a half to two times the runtime. Title episodes with the guest's most provocative claim, never 'Episode 4 with Jane Smith'; the title is doing the marketing, the numbering is doing nothing.

Repurposing: one recording, ten assets

A branded podcast on a small budget only pays off if every recording feeds every other channel. The trick is to mark the moments live: keep a notepad open during recording and jot a timestamp whenever a guest says something spiky. Those notes become your clip list, and transcript-to-clip tooling — a platform like AI BOSS among them — can turn the marked moments into captioned vertical clips in a single batch pass.

  • Two or three captioned vertical clips of 60 to 90 seconds for Reels, TikTok, and Shorts.
  • One audiogram of the single best exchange for feed posts.
  • Two or three quote graphics pulled straight from the transcript.
  • A newsletter section summarizing the episode's one big idea.
  • A blog post edited from the transcript, targeting the question the episode answers.
  • A LinkedIn post built on the guest's spikiest opinion, tagged so they reshare it.

Growth expectations: what good actually looks like early

Here is the honest part. Early listens come almost entirely from your existing email list, your team, and guests sharing their own episodes. Growth is lumpy, not linear — it moves when a well-networked guest shares, when a clip travels, when an episode nails a searched question, and it flatlines in between. Judge season one on qualitative signals instead: are guests saying yes faster, are prospects referencing episodes unprompted, are newsletter replies mentioning the show. Then hold the episode-ten review and ask three questions: did production get easier each episode, did anyone act because of the show, and do we have ten more episode ideas we are excited about. Three yeses means renew. Anything less means fix the weak answer before recording episode eleven.

A podcast is not a growth channel in its first season. It is a relationship channel that happens to publish audio.

One last piece of insurance: batch-record the first three episodes before you announce anything. Launching with a small library protects your cadence through the inevitable chaotic weeks, gives new listeners something to binge, and — most usefully — forces you to survive the production process three times before you have promised anyone a season.