Managing Podcast Production Without an In-House Team

Running a podcast in-house looks like one job. It is about six, and five of them are invisible until they are late.

Three things here, in the order you will need them: what the work actually is and whether to keep it, the workflow that makes it ship, and how to diagnose it when it stalls.

The six jobs behind one episode

• Guest pipeline. Finding, inviting, chasing, rescheduling. Never finished, and the first thing to lapse.
• Scheduling and logistics. Calendars across time zones, links, briefs, reminders.
• Recording support. Levels, backups, the guest whose microphone is the laptop.
• The edit. Content first, then audio, then video.
• Packaging. Show notes, chapters, thumbnail, metadata, clips.
• Distribution. Publishing, scheduling social, the newsletter, and actually getting it in front of non-subscribers.

Most in-house shows staff the middle two and quietly hope the other four happen. They do, for about five episodes.

What in-house really costs

• Attention, not hours. The work is spiky and interrupt-driven. It fragments the day of whoever owns it.
• Single-point-of-failure risk. One person usually holds the whole workflow. When they take leave or resign, the show stops.
• The opportunity cost of a marketer. The person capable of running this well is capable of running your demand gen.
• Tooling and kit. Recording platform, editing software, hosting, design, scheduling. Individually small, collectively not.
• Quality drift. Without a producer, standards move slowly downward and nobody notices until an episode is genuinely bad.

Keep it in-house when

• Someone genuinely wants the job, and it is in their objectives rather than added to them.
• The format is simple — audio-only, one recurring host, few guests.
• You are still figuring out what the show is, and want to learn fast by doing it cheaply.
• Your cadence is monthly. The load is manageable at that pace.

Outsource when

• You have missed two publish dates in a row. That is a system failure, not a discipline failure.
• The show is video, or you want clips. The workload roughly doubles.
• The owner is your only marketer, and the podcast is eating the rest of their remit.
• The show matters commercially. Once pipeline depends on it, hobby-grade reliability stops being acceptable.

The middle option most teams miss

Keep the relationships. Outsource the pipeline. You keep the guest asks, the point of view and the promotion post — the parts where your name and judgement matter. A partner takes scheduling, recording support, editing, packaging, publishing and distribution — the parts that are pure throughput.

The workflow that ships

A podcast that ships reliably is not a discipline achievement. It is a workflow with named owners and fixed dates, written down somewhere other than one person's head. This is the one we run, and it is entirely runnable in-house.

Work in fortnights, not episodes. One fortnight contains a record, a publish, and preparation for the next, running in parallel. Teams that treat each episode as a project starting when the last one finishes are always one disruption from a gap.

Stage 1 — Guest pipeline (continuous)

Owner: the host. A standing list of twenty names, ranked, each with who owns the introduction. Refill monthly. The invitation comes from the host; the chasing does not have to. Shows fail here more often than anywhere else, and silently — there is no missed deadline, just an empty calendar in three weeks.

Stage 2 — Booking and brief (T-14 days)

Owner: producer or coordinator. Calendar hold, recording link, and a one-page brief to the guest: format, length, themes, what to prepare, technical requirements. Ask for a headshot and preferred title now, not the day you publish. The brief is the most useful document in the workflow — a prepared guest is a better episode and a shorter edit.

Stage 3 — Prep (T-2 days)

Owner: the host. Twenty minutes reading, and three questions nobody else would ask. Not a script, a spine. If the host cannot name the one thing they want out of the conversation, the edit will not find it either.

Stage 4 — Record (T-0)

Owner: producer. Join early, check levels, confirm separate tracks per speaker, start a local backup. Ten minutes of setup discipline saves hours in post, and mic technique matters more than gear — the short version is here.

Stage 5 — Edit and package (T+1 to T+5)

Owner: editor. Content edit first: what stays, what order. Then audio, then video, then the assets — show notes, chapters, thumbnail, metadata, clips. Package as one job rather than trickling assets out over a week.

Stage 6 — Review (T+6, one window)

Owner: one named reviewer. A fixed window, one round, one consolidated set of notes. Open-ended review is the most common place a workflow stalls, because there is no definition of done.

Stage 7 — Publish and distribute (T+8)

Owner: producer. Feed, YouTube, directories, then the distribution work: host's post, clips scheduled across the fortnight, newsletter slot, guest sent their assets. Publishing takes an hour; distribution is the fortnight. The guest handoff is the cheapest growth in the whole workflow and the most often skipped — send them the clips, the link and a suggested caption.

The rule that keeps it alive

Never let the buffer fall below two recorded episodes. Everything else is preference; this one is structural. Two banked episodes absorb a cancellation, a sick week, a launch and a board meeting. At zero buffer every disruption becomes a missed publish date, and two missed dates is how shows quietly end. Record in blocks to build it.

When it stalls: the four bottlenecks

A stalling podcast almost never has four problems. It has one, and the symptoms make it look like a general lack of time. Diagnose before you fix.

1. The editing backlog

Symptom: recordings exist that have not been published, and the record-to-publish gap is growing. Cause: editing is done in leftover time by someone whose main job is something else, so it gets deferred. Fix: make it scheduled work or move it outside. This is the one bottleneck money reliably solves, because the edit is genuinely separable — see what a professional edit covers.

2. Guests and scheduling

Symptom: nothing in the calendar three weeks out. Cause: booking is reactive, and good guests need more notice. Fix: the standing list of twenty, refilled monthly, and batch-record to build a buffer.

3. Review and approvals

Symptom: episodes sit "nearly done" for a week, with fragmentary and sometimes contradictory feedback. Cause: no definition of done and no single reviewer. Fix: one named reviewer, one fixed window, one round. Agree what a revision is for — factual and reputational issues, not taste.

4. Promotion falls off

Symptom: episodes publish on time and nothing happens. Cause: the workflow ends at publish, so distribution is whatever energy is left. Fix: treat distribution as a scheduled stage with its own owner and checklist.

Which one do you have?

• Recordings piling up unpublished → editing.
• Empty calendar three weeks out → guests.
• Episodes stuck at ninety per cent → review.
• Shipping on time, flat results → promotion.

If two or more are true at once, the problem is capacity rather than process, and adding steps will not help. Every one of the four is the same failure underneath: a stage with no named owner and no fixed date.

What handing it over costs

$2,450 +GST a month covers two fully produced episodes with the whole loop above handled, $4,900 +GST for four. A one-off $2,000 setup applies to an existing show, waived if you develop it with us, and full show development is around $18,600 +GST. A fuller comparison of providers is in the buyer's guide.

Frequently asked questions

What makes in-house podcast production so hard?

It is six jobs disguised as one, and five are invisible until a deadline is missed. The recurring, unglamorous work — guests, scheduling, distribution — decides whether a show survives.

Can one person run a podcast alongside another role?

At monthly cadence, usually. At fortnightly with video, rarely for long. The failure is gradual: publishing slips a few days, then a week, then the show is dormant.

How long does one episode take end to end?

Elapsed, about eight days from record to publish on this workflow. Hands-on time is much smaller — the calendar is mostly buffer, deliberately.

We have all four bottlenecks. Where do we start?

Guests, then editing. An empty calendar stops the show entirely; a backlog only delays it. Fixing promotion first feels productive and changes nothing while the pipeline is dry.

Will hiring an editor fix a stalling show?

Only if editing is the actual bottleneck. If the problem is guests or review, an editor sits idle waiting for files.

Where to start

Write the seven stages down this week and put a name against each. The empty rows are the whole diagnosis.

If it turns out to be capacity rather than process, thirty minutes on a call will tell you what handing over the loop costs. No deck, no pitch.

Book a call with Adam.