One export spec, not five different exports
This page holds the export settings, not the release-day order. The release day checklist has that order, and editing software for beginners covers which app cuts the episode. What follows: loudness, sample rate, bit depth, channels, format, and the check to run before a file ships.
The loudness target every platform ends up honoring
Apple Podcasts publishes an exact number: -16 LUFS integrated, plus or minus 1 dB, with true peak under -1 dBFS. Spotify and YouTube do not publish a matching spec for podcast delivery; both normalize loudness on playback instead, turning a hot file down and leaving a quiet one alone. Build to Apple's number and the other two follow.
Apple's own documentation states this target in LKFS, the older name for the same loudness standard as LUFS. The two units are interchangeable. A meter reading either one against -16 is reading the same target.
| Platform | What it does with loudness | What you export to |
|---|---|---|
| Apple Podcasts | Publishes a fixed target: -16 LUFS integrated (plus or minus 1 dB), true peak under -1 dBFS | Build your master to this number directly |
| Spotify | Applies its own loudness normalization on playback; no published fixed spec for podcast delivery | The same -16 LUFS master carries through without a second pass |
| YouTube | Normalizes on playback the same way, turning a hot upload down and leaving a quiet one untouched | The same master avoids a turned-down, thin-sounding episode |
Sample rate and bit depth, from mic to master
Three stages, three settings. Record at whatever your interface supports, at 24-bit resolution. Keep your working master at 44.1 kHz for audio-only shows, or 48 kHz for an episode recorded alongside video, matching the frame rate without a resample. Bit depth stops mattering once the file lands on a lossy delivery format.
| Stage | Sample rate | Bit depth | Note |
|---|---|---|---|
| Recording | 44.1 kHz or 48 kHz (48 kHz when video is involved) | 24-bit | Extra headroom for noise reduction and leveling later |
| Working master (WAV) | Matches your recording rate | 24-bit | Matches Apple's recommended resolution for a direct-upload WAV or FLAC master |
| Final delivery (MP3 or AAC) | 44.1 kHz or 48 kHz | Not applicable, a lossy format | Matches the rates Apple and most RSS hosts accept |
Mono or stereo, and what actually changes the file
A single mic, one voice, dialogue only: export mono. The file drops to about half the size, and there is no stereo image to lose. Multiple mics, a co-host panned apart, or a music bed under the intro: export stereo. A fold-down to mono collapses anything panned and can thin the mix.
- Apple's RSS delivery accepts mono at 64 to 128 kbps and stereo at 128 to 256 kbps, both at 44.1 or 48 kHz.
- Apple Podcasts Connect, the direct-upload path, requires a stereo file. If your source is mono, duplicate it to both channels rather than leave one silent.
- There is no penalty for choosing mono on a voice-only show. The smaller file uploads faster and re-encodes cleaner at every step downstream.
The format and bitrate you actually ship
MP3 stays the practical default. Every podcast app and every host decodes it without a plugin. Apple's RSS guidance accepts MP3 or AAC, and recommends AAC for a cleaner file at the same bitrate where your host supports it. The bitrate table below is what actually ships.
| Format | Bitrate | Sample rate | Fits |
|---|---|---|---|
| MP3, mono | 64 to 128 kbps | 44.1 or 48 kHz | Single-mic, voice-only shows |
| MP3, stereo | 128 to 256 kbps | 44.1 or 48 kHz | Multi-voice shows or a music bed |
| AAC | Apple's recommended format at the same bitrate range | 44.1 or 48 kHz | Hosts and RSS feeds that accept it |
The pre-publish QC pass
Run this on the final export, not the working session file. It takes five minutes, and it catches nearly everything a listener or a sponsor would otherwise catch first. Work the list in order and stop only when every line checks out clean.
- Measure the integrated loudness on the exported file itself, not the project inside your editor. -16 LUFS, plus or minus 1 dB, is the number to land on.
- Check the true peak reads under -1 dBFS. A limiter set right at 0 clips the moment a phone or a podcast app applies its own gain.
- Listen to the first thirty seconds on a phone speaker, not studio monitors. Most of your audience never puts on headphones.
- If you exported stereo, fold it down to mono and listen again. Anything that thins out or disappears is a phase problem worth catching now, not after a listener flags it.
- Confirm the ID3 tags: episode title matches the RSS title exactly, episode number is set, and artwork is embedded, not just attached in the host's dashboard.
- Check the file boundaries for a click or a breath cut off at the very start or the very last second.
- Play the exported file start to finish at real speed once. A dropped export, a doubled section, or a wrong file length only shows up this way.
Where the spec lives, to stop it from drifting
A number written down once stays the same at episode sixty. A number someone remembers gets rounded differently every few weeks, and nobody notices until a sponsor's audio engineer flags a mismatch between two ad placements recorded months apart.
Pod Green Room stores the export spec on the same podcast CRM record as the episode. Pod Green Room keeps it next to the episode's sponsor tracking, and a media kit built later pulls the same loudness number without a second look.
What this checklist does not decide
This is the audio spec, not the whole release. Chapters, show notes, artwork updates, and the sequence of who gets emailed when belong to release day, not to the export settings on this page. Get the numbers right here and the rest has a clean file to work with.
Lock these numbers once, per show, and revisit them only when something upstream changes: a new microphone, added video, a co-host who records from a different room. The export spec is the one part of the episode meant to stay identical, week over week.