Recording Podcasts at Home: Equipment, Software, and Common Mistakes

After engineering more than 1,400 podcast episodes across business, true crime, and narrative formats, I have a working answer to "what do I need to start a podcast." It is much smaller than the lists you'll find on most blogs. It's also more specific.

The minimum setup that actually works

Here is a complete podcast recording rig that has produced shows in the iTunes top 200:

• Mic: Samson Q2U or Audio-Technica ATR2100x ($75–$100)
• Headphones: Sony MDR-7506 ($100). Closed-back, comfortable.
• Boom arm: Cheap desktop boom ($30)
• Pop filter: A foam windscreen on the mic, $10
• Recording software: Audacity or Reaper. Free or $60.

Total: about $215. Anything beyond this gets you marginal improvements that most listeners will not consciously perceive. The Shure SM7B + cleanlink + nice interface stack ($800+) sounds noticeably better to engineers and barely different to audiences.

Microphone: the one that matters

For podcasting, the microphone choice is more constrained than for music. You want a dynamic mic with a tight cardioid pattern that rejects room sound. The reasons:

1. Your room is untreated.
2. You will record next to a keyboard and a screen with cooling fans.
3. A condenser will hear all of that.
4. You don't need condenser detail — you need rejection.

Dynamic mics built for voice (the SM58, SM7B, Q2U, ATR2100x, RE20, Procaster) are designed to make this trade. More on the dynamic vs. condenser decision here.

One technique note: get close. Two to four inches from the mic, not eight. Close mic technique is what produces the intimate, "in your ear" sound listeners associate with professional podcasts.

Software: recording & editing

For recording solo: Audacity (free) is more than enough. Reaper is better when you start handling multi-track edits.

For editing: Descript has changed the workflow for many podcasters because it lets you edit audio by editing a transcript. For long-form conversational shows, this can cut edit time by 40%. The catch: it can introduce subtle audio artifacts that purists find unacceptable.

For mixing finished episodes: A simple chain of high-pass filter, light compression, and a brick-wall limiter will get a voice to broadcast-loudness. Modern podcast platforms target -16 LUFS integrated loudness; aim for that.

Recording remote guests

This is where most podcasts go wrong. The temptation is to "just record the Zoom." Don't.

Zoom (and Google Meet, and most video conferencing) heavily compresses audio for stability over poor connections. The result is fine for meetings, terrible for podcasting. The standard solution is one of:

• Riverside.fm or SquadCast. Records each participant locally on their own machine, then uploads after the call. Result: studio-quality audio for everyone, regardless of connection. $20-$40/month.
• Double-ender. Each participant records themselves locally to a DAW or to Voice Memos, then sends the file. Works fine, costs nothing, requires technical comfort from your guests.

Skip platform recording. The audio difference is the single biggest perceived-quality lever you have.

The four most common mistakes

1. Mic too far away. New podcasters subconsciously treat the mic like a conversation partner and back off. Get within four inches. Look at how a radio host sits at a microphone.

2. Inconsistent levels between takes/episodes. Set a single input level the first time and write it down. Use it every episode. Listeners adjust to a level; jumping around between episodes feels amateur.

3. Over-editing. First-time editors remove every "um," every breath, every pause. The result sounds robotic. Leave the human texture in. Edit out the parts that don't serve the story; preserve the parts that make it feel like a conversation.

4. Skipping the loudness normalization step. An episode mixed at -22 LUFS will sound thin and quiet on Spotify next to other shows. Master to -16 LUFS for podcast platforms, or use a loudness analyzer plugin (free options exist) to check.

An episode workflow that scales

1. Pre-record: Send the guest a one-pager with mic placement guidance, headphone reminder, "no carpet on floor, please use it" instructions.
2. Record: Local audio capture on both ends. Backup recording on the host side.
3. Rough edit: Remove false starts, long silences, and obvious mistakes. Don't tighten yet.
4. Story edit: Listen front to back. Cut anything that doesn't move the conversation forward. This is the longest step.
5. Mix: HPF, compression, limiter. -16 LUFS target.
6. Export: MP3 96 kbps mono or 128 kbps stereo for conversational content. Higher bitrate for music-heavy shows.
7. Publish: Upload, write show notes, schedule.

The shows that succeed long-term are not the ones with the most expensive microphones. They're the ones that publish consistently and don't make listeners work to understand them. Eliminating background noise is one of the cheapest ways to make that easier.