Podcast Memorandum, or A Brief Guide for Aspiring Podcasters (Updated Edition)

Oct 13, 2021

22 min

miscellaneous

Version 0.5.0.

what is this?

This document reflects the purely subjective opinion of myself — a podcaster and podcast listener with 15 years of experience and many hundreds of recorded podcasts. The brevity of the format was chosen deliberately, in the hope that you won’t get tired while absorbing all of this and will perhaps even read it to the end.

can I become a famous and listenable podcaster?

You have to accept that not everyone can do this, even if they really want to. There are quite specific contraindications for the spoken genre that, in my opinion, are perfectly obvious. These include significant speech defects, inability to speak publicly, inability to hold a thought, various accents, and a whole range of other equally understandable criteria.

To compensate for the previous point — a fairly significant percentage of people can do it from birth, and an even larger percentage can learn. There’s no particularly complicated science here. Recording a podcast that’s interesting and pleasant to listen to and that will certainly find its audience — that’s entirely within your power.

Among podcasts there’s a variety of types, and you’d probably do well to choose the most suitable one and develop in that field. Here’s a brief list of categories:

Mono-podcasts (monologues) — suited for confident, mature, and experienced individuals. The hardest genre for beginners. In very rare cases, a very young person can pull it off, though maybe you are that exception.
Topic-based dialogues (two participants) — perhaps one of the simplest genres, good for virtually all categories. Recommended by me as the default for beginning podcasters. The pairing of conversation partners is very important — temperament, like-mindedness, rhythm, and some mystical parameter of “thinking on the same wavelength.”
Free-form dialogues — better not to attempt unless you’re already a star. But if you’re already seasoned, you don’t need my advice.
Podcasts with more than two participants on set topics — noticeably harder than a simple dialogue, though quite suitable for advanced beginners. One of the podcasters must be the host, and different requirements apply to that role. With a host capable of keeping the conversation on track, participation of people with different opinions and different thinking speeds becomes possible.
Interviews — this type of podcast is quite accessible to beginners, not much harder than a dialogue, and possibly easier. There’s somewhat less spontaneous live interaction in this genre, but the importance of show preparation increases significantly.

what to talk about?

You can talk about anything in podcasts, as long as the topic interests you and you feel you have something to say about it.
You don’t need to be an expert in the field. Your interest in the topic and the ongoing process of deepening/broadening your understanding in the chosen area is quite sufficient. For example, a podcast by someone discovering something new in each episode can be no less interesting than a podcast by a guru.
Within your topic, you should have enough material for at least 10 podcasts ahead.
You can talk about what moved you (within your chosen topics) since the last episode. This is one of the most accessible methods for finding new stories, though you need to be immersed in the field for this to work.
Very narrow topics can be equally interesting as extremely broad ones. For example, you can successfully talk about the specifics of ice fishing or about questions of human philosophy.
Mixing completely different directions in one podcast is not the best idea. However, the breadth of your chosen area is defined by you and can cover many loosely related topics.

how to speak well?

There are methods that can improve speech technique. Some materials you can find here and here, and Google will help you find even more.
Practice, practice, and more practice. Speak into a microphone a lot, often, and for long periods. Eventually, you’ll most likely be able to do it well.
Always, meaning always without exception, listen to yourself through headphones. If your equipment allows it, during the conversation (preferred) or after.
Re-listen to your podcasts, even if it makes you cringe. This will pass with time, but by re-listening in a calm environment, you’ll certainly notice flaws — both in content and in speaking style.
Write down your frequent mistakes and keep this shameful list in front of your eyes during podcast recording.
You should have many words in your internal vocabulary. The best way to expand your personal vocabulary is reading quality texts. Modern and classic literature will help you. In other words — read books.
While reading, pay attention to what strikes you as beautiful. Write down these phrases, re-read them, and say them aloud.
Listen to those whose speech you admire. Even better — talk with these people.
If you’re single, find a partner smarter than you and talk with them a lot. This is even better than reading books :)
Accept that you won’t be perfect. And you don’t need to be perfect — we’re not trying to become professional announcers. Ums, ers, and filler words are not an automatic sign of “ugly” speech, as long as they don’t overflow it.
Roll up your sleeves and carefully go through the podcast, removing everything that makes you cringe. For example, nose-blowing is a good candidate for removal. Excessively frequent filler words and flubs are also worth cleaning up, but don’t go overboard.

preparing for a podcast

Always prepare for podcasts, even if it’s pure improvisation.
All podcasters always prepare for their episodes, even the most accomplished, relaxed, and talkative ones. If this idea hasn’t reached you yet, re-read the previous point.
Podcast preparation varies significantly depending on the genre and topic. At minimum, you need to come up with at least a rough outline of what you’ll talk about / what topics to discuss / what questions to ask.
Write down everything you’ve come up with. It doesn’t matter where — what matters is not keeping it all in your head.
Don’t go crazy and prepare long cheat sheets. Write just enough to remember during the podcast what it’s all about. Sometimes a word is enough, sometimes a brief sentence.
Start preparing for the next podcast immediately after publishing the previous one. Don’t put it off until the last day.
Come up with a preparation technique that works for you — convenient and accessible anytime, anywhere. For example, a voice recorder or notebook where you can jot down a hint about a topic while it’s fresh in your mind.
Right before recording, process the accumulated material. Sort it in the order you’ll discuss it. If you plan to respond to comments and questions, add those to your materials too.
If you use audio materials or production elements in your podcast, prepare them in advance. It’s worth organizing a small library of sounds/effects in a format that makes sense to you and keeping it handy.

how to record properly?

A beginning podcaster typically doesn’t have the means or desire to buy relatively expensive equipment. However, perfectly accessible and affordable devices will let you record something that can be listened to comfortably. The main goal of the entire “proper recording” process is not to sound like a top FM radio station, but to be listenable.
As a minimally sufficient recording device, consider various USB microphones, as well as inexpensive portable recording devices like the Zoom H1n or the slightly more expensive Zoom H2n.
Before using your microphone/headset/recorder, learn how to speak into it properly. Experimentally determine the optimal distance and position.
In the vast majority of cases, a minor modification by adding a DIY (wrap the sound receptor in foam) or store-bought windscreen will eliminate or reduce defects with “p” and “b” sounds (so-called plosives).
Recording can be done in any program that allows recording an uncompressed audio file (wav, aiff, etc.).
Minimal audio post-processing, especially when recording on simple equipment, is a mandatory part of the process.
As recording and editing software, you can use the free Audacity or inexpensive comparable programs (SoundStudio, GarageBand, Ubercaster, Sound Forge, RecordForAll).
The recording level while speaking should not exceed –3dB at the loudest points. For greater safety, try to keep it below –6dB to avoid clipping (exceeding 0dB), which sounds very unpleasant and can’t be fixed in post-processing.
Audio should be processed with 4 filters: Expander (Noise Gate), Compressor, Limiter (or Compressor in Limiter mode), and Normalizer.
You should at least roughly understand what these processes do. For practical-theoretical materials, see Appendix 2.
Don’t try various “cool effects” if you don’t really understand what they are or how they work. There shouldn’t be many effects. Less is better than more.
Properly processed audio will sound at a nearly constant, close-to-maximum volume throughout. By ear, the difference between the loudest and quietest moments shouldn’t exceed 20-50%. The smaller the better.

what not to do?

Don’t record a podcast if you have nothing to say.
Don’t force yourself to record podcasts.
Don’t take long breaks between podcasts. After a month-long break, it probably makes more sense to start a new podcast and maintain regularity in the new one.
Don’t record very long podcasts. A podcast longer than 35-45 minutes, especially from a beginning podcaster, will very rarely be listened to in full.
Don’t record very short podcasts. A podcast shorter than 5 minutes is suspicious, and shorter than a minute is alarming.
Don’t brag about poor audio quality in podcasts, justifying it with good content.
Don’t boast about good audio quality in podcasts or talk at length about how you achieved it.
Don’t start your podcast by explaining “what podcasts are.”
Talking simultaneously with other podcast participants is a harmful practice — it’s generally very hard to follow when listening.
Don’t try to replicate some other podcast, especially a popular one.
Don’t force-decorate your delivery with profanity.
Never read text aloud, even if it’s yours. Speak, but don’t read.
Don’t try to speak more beautifully/smartly than you do in everyday life. It will be noticeable and amusing. Avoid using words whose meaning you’re not entirely sure of.

In podcasts, the social element matters, though not as much as in blogs. However, feedback (comments, interaction on Twitter/Telegram/Facebook, direct correspondence) is a very important part of the podcast product you produce.
Allow the most open comments possible on your podcasts. Don’t force listeners to register just to comment.
Use comments in the podcast. Not all of them — only those you want to respond to. But don’t turn the entire episode into a discussion of comments. 10-15% of the time is quite sufficient.
Accept that not everyone loves you. At the same time, try to understand where people simply don’t like you versus where there’s a hint of constructiveness. In the first case, don’t try to argue/convince; in the second case, try to perceive the criticism as objectively as possible.
Don’t feed the trolls and ruthlessly delete and block rude people.
Ask those whose opinion matters to you. They usually won’t say anything on their own but can answer a direct question.
Regardless of where your podcast lives originally, try to publish it in places where listeners congregate (directories and aggregators). Always include a link to the primary location.
Once you roughly understand the size of your podcast audience, reconsider the relevance of both positive and negative comments. Remember that audience size is the one and only indicator of podcast success, and that those who like everything say far less than those who don’t like you or your podcast.
Many comments, even bad ones, are better than none at all.

appendices

frequently asked questions

How to maintain the drive for podcasting? Simply put, what makes you sit down at the microphone every week?

Generally, you shouldn’t force yourself to sit at the microphone if you don’t want to. And if you did force yourself but it’s not flowing — drop it. Without the “drive” it won’t work or won’t go as you’d like. However, if this is truly your hobby, you won’t need to force yourself. It’s also worth analyzing what’s stopping you. If you simply don’t feel like it, don’t torture yourself and your listeners. If you’re short on time — develop more optimal recording and processing techniques. For example, I spent no less than 3-4 hours of meticulous work on the first 50 Radio-T podcasts, but now after about 15 minutes of pre-preparation, everything happens on its own and takes another 10 minutes for uploading and writing show notes.
What equipment is needed for decent audio quality but at an acceptable price (we’re students) — the basic minimum?

I consider the $65 Samson Q2U to be minimally decent equipment at a modest price. People successfully order it from eBay and Amazon. If you have a bigger budget, look at the numerous USB microphones in the ~$100 range, such as Rode NT USB-Mini, Elgato Wave:3, Audio-Technica ATR2100X-USB, Audio-Technica AT2020USB, and many others. Another acceptable option in the “under $150” range is the Zoom H2N.
How to organize a podcast with multiple participants (what does each participant need to do)?

This depends on the podcast type. In dialogues, participants are typically equal conversation partners and converse in the most ordinary way. Pay attention to avoiding long pauses, but in all other aspects it’s no different from a normal conversation. A “three-way” podcast is a completely different beast. Here the host is crucial — they lead the conversation in the intended direction, watch for long monologues, and engage all participants. Participants, of course, should help with this. It’s worth agreeing that when the host starts speaking, everyone else instantly goes silent. It’s also worth agreeing not to start speaking while another participant is talking. Stopping someone who’s gotten carried away is the host’s job.
Background music: Technical effect, decoration, or bad taste?

Background music can be both a “feature” and a technical means of masking background noise. As decoration it’s more dubious than useful, but as a masking tool — sometimes indispensable. Background music, even the most monotonous, is perceived better than plain noise, especially irregular noise.
What to use for recording and editing podcasts? Which programs?

On this topic, it’s worth listening to the “Brief Comparative Review of Recording and Editing Programs”.
How to check if you’ve compressed the audio correctly? How to know that the slider positions you’ve chosen are right? Is over-compressing scary?

The simplest way is to compare a section of your audio with a section of someone whose compression you like. It’s important to compare pieces of roughly the same length at roughly the same zoom level. We’re talking about visual comparison in audio editors. Over time you’ll learn to perceive this by ear, and your ears will become the primary measure of compression. Over-compression, as long as it doesn’t hurt your ears, isn’t very scary.
How to record the third podcast episode? :) Because it often happens that you record the first episode — nobody listens. Record the second — again nobody. Then you just lose motivation.

Here the question is probably more about why nobody’s listening and when to decide “it didn’t work” and call it quits. There’s no universal answer. If you think the podcast is good in content and decent in sound, maybe it just hasn’t been noticed. Tell everyone you can about it. Try telling podcast authorities about it — if they like it, they’ll help. Also, the question is — what is “nobody”? For certain niches, even 100 people is a lot, and for others, even 10,000 is too few. Recording a podcast about audio engineering, I’m not at all surprised that only 1,500 people listened, since the field is narrow and by my expectations even 500 would be a lot. Of course, if you’re unhappy with your audience size or you observe a clear downward trend from episode to episode, it might be worth giving up. After the 3rd episode it’s probably too early, but after the 10th you can consider it.
What to choose — a cheap USB microphone or a cheap regular dynamic microphone + preamp?

In terms of price/quality, a USB microphone is better than a dynamic with a preamp.
How to export to mp3 from Audacity so it doesn’t re-compress the audio? Because in the program itself everything sounds fine, but after rendering…

It won’t ruin anything beyond what you asked for. Choose sample rate 44.1 and bitrate from 64k, and there won’t be any noticeable degradation.
How to speed up podcast editing as much as possible?

The most effective way is to eliminate the editing stage entirely. That is, move post-processing into processing during recording and record podcasts clean from the start, without editing the result. Processing requires equipment that isn’t the cheapest (vocal strip, $250-$800), and “clean speech” requires serious podcasting skills. The most accessible decent vocal strip is the DBX 286s.

Another useful trick is to place markers during recording at points where you made a mistake that’s worth fixing. Some recording and editing programs have built-in tools for this, but the most universal solution is a “dog clicker” (almost free on Amazon). On the recording, the click will be very distinctly visible, and you’ll always know which spot needs processing/cutting. With this system, you click when something goes wrong (forgot a word, said something wrong, paused too long), wait a couple of seconds, and re-speak the failed fragment.
Recording online conversations. The topic is too broad for an FAQ, but at least point me in the right direction (software names, routing methods).

This is apparently about recording online calls. There are 2 recording methods: software and hardware. Software is simpler and cheaper, hardware is more universal and reliable. For macOS, Audio Hijack does this beautifully, and for Windows — MP3 Skype Recorder. Doing it with hardware is harder — you need input/output splitting, blocking of one stereo channel, plus a second audio card or external recorder. However, if you have something like the RODECaster Pro or Zoom PodTrak, this is supported out of the box.
How to properly structure the presentation?

Presentation structure is a rather murky subject. Don’t over-formalize the process, though it makes sense to come up with a basic skeleton and build everything around it. For example: greeting, a couple of minutes warming up yourself and the audience with references to past episodes, going through topics (ideally with thought-out transitions between them, though not mandatory), reviewing comments and replies, and a nice sign-off. The structure can be reversed too, but it’s good to have one, at least roughly and flexibly. Listeners will know what to expect from you, and met expectations will please them.
You keep talking about compression. What is it? How to properly make mp3 files?

No, I’m talking about basic dynamic processing. Audio compression, to put it simply, makes loud moments quieter and quiet moments louder. It narrows the dynamic range, averaging out your sound. With such a seemingly straightforward function, it’s hard to believe how much proper compression can improve sound and provide listeners with comfortable and pleasant audio perception.
How often should you record podcasts?

There’s no universal rule here. For very short podcasts, daily can be normal; for half-hour ones — once a week. Don’t do it too rarely — in my opinion, once every 2 weeks is the absolute limit.
I can’t edit yet, I don’t have a microphone, but I want to record. Can I?

You can, but better not. At minimum, something to speak into (see above) and basic post-processing knowledge are worth acquiring.
How many podcasts to record in one sitting? I heard the recommendation is to do them in batches.

One podcast per sitting. With “batch” recording you lose the feedback loop, and it’s more of an industrial approach than a creative one.

some basics of audio processing

All post-processing — operations performed after recording — is done at the very last stage. If you’re doing editing, cutting and trimming things, that should be done before applying any filters. In all modern audio editors, there’s a menu item usually called “Filters.” That’s where all the magic hides. Basic audio processing includes several mandatory and several optional stages. One of the fundamental principles states: “First remove, then add.” That is, start with what removes something from your sound.

From the whole variety, what matters to us now is Expander/Dynamic Expander. This filter allows you to noticeably improve the sound in places that, while quiet, are unnecessary. We’re talking about noise, breathing, lip smacking, and other completely superfluous sounds. The Expander removes them by setting just two main parameters: Threshold and Ratio. Threshold, in this and all other filters, defines the trigger level. Ratio defines the degree of effect application. This is also a common concept for many filters, but in this case it means how strongly to suppress sound that’s quieter than the set Threshold. No universal rules or correct settings exist. It depends on the specifics of your audio, its level, and the nature of the unwanted sounds. The only recommendation is don’t overdo it. This filter, despite only processing quiet sounds, can surprisingly and disgracefully worsen your audio. As a rule, don’t choose too high a Ratio. Up to 1.5 is usually sufficient and safe in practice. Threshold between –30 to –40 dB can be used as the recommended range, assuming you followed the above level recommendations during recording.

After quiet but unnecessary sounds are eliminated, you can tackle the main, loud part. Our task is to make it more even in level and louder overall. This is necessary for comfortable listening. If the Expander can sometimes be skipped, especially if the audio is clean in quiet spots, the Compressor can never be ignored. In some programs it may be called Dynamics Compressor or simply Dynamics, and sometimes it’s combined with the Expander. In terms of parameters, the compressor is very similar to the Expander. The two mandatory parameters are also called Threshold and Ratio. However, the meaning of Threshold here is directly opposite — it defines the level (volume) above which everything louder will be attenuated. As you might guess, Ratio sets the degree of this attenuation. So, for example, Ratio=2 means everything above Threshold becomes 2x quieter. The result of applying the compressor is a reduction in volume of those moments you’ve defined as loud. Obviously, defined by setting the appropriate Threshold for your specific case. What’s appropriate — only you can answer.

Look at your file in the audio editor and decide what your average level is. Then reduce that number by about 30% to get your Threshold. Ratio you can choose to taste. There are fans of aggressive compression (Ratio 3-6 or even higher) and low compression (less than 3). Experiment, listen, and decide what suits your taste.

After the compressor, it’s sometimes worth running a Limiter/Leveler filter. Essentially it’s just a Compressor with the highest possible Ratio and the shortest possible Attack Time. It’s used for removing peaks. As a rule, beginners have peaks. Looking at your audio, you’ll easily spot their presence or absence. If they exist, set the Ratio to the level above which you don’t want to hear anything. That is, everything louder will be attenuated to the Threshold level.

A couple more parameters not yet clearly discussed are Attack Time and Release Time. The first determines how long the signal level must be quieter (for Expander) or louder (for Compressor/Limiter) before the filter kicks in. Release Time, as you’ve probably guessed, is the opposite — it means how long the signal level stays louder (Expander) or quieter (Compressor) before the filter stops acting. These parameters are worth playing with, but within reason. Usually for voice recording, Attack Time is chosen between 0.010-0.050 sec, and Release Time between 0.05-0.150 sec, depending on your voice and what sounds good to your ears.

The last mandatory stage is Normalization (Normalize). It raises all your audio to a set maximum. Usually you set just one parameter — Level. Either in percentages or negative dB values. It should be set close to 100% or 0 dB, but slightly below maximum. 95-97% is the correct value (–0.4 to –0.1 dB).

simplified podcast production process

This section explains the process that I see as the simplest and most time-efficient. The simplicity of the production process is a very important factor that can be the very thing that determines the difference between “maybe I should record a podcast now” and “forget it, too much hassle and no time to bother.”

The essence of the process is that we try to avoid fully re-listening to recorded material, turn the cleanup process into something that takes a couple of minutes, avoid constantly configuring and reconfiguring filters, and ideally avoid any post-processing on the computer entirely. This process, as I use it, is somewhat more expensive in initial investment, but saved time isn’t free either, and it will save you a lot of time.

Record the podcast from start to finish without stopping, as if it were a live broadcast. Ideally, all you need to do is record properly the first time. However, this rarely happens, and for such cases use the dog clicker (see above).
During editing, find the spots with vertical lines on the recording (they’re very visible, impossible to miss) and cut out the bad fragment.
If possible, apply effects during recording itself. This means either a vocal strip (see above) or modern USB microphones that come with software allowing this on the fly. Such software exists, for example, for Shure, RODE, and some other microphones. In this case you won’t need software filters at all, and the result (after necessary cuts) can be saved directly to mp3 and published.
If that’s not possible and effects can’t be applied during recording, use a DAW with preset/template support. One such product is Reaper — you can configure all effects once, save as a template, and after that all settings including effects will be applied to new recordings automatically. Essentially, this is another variant of “applying effects live,” but unlike the previous point, the effects here are software-based, not hardware.
If your podcast includes materials besides your voice, like intros or bells during the show, find a way to include them during recording. There are several products for this, like Soundboard Studio and Podcast Soundboard, that let you play pre-recorded sounds directly into your podcast.
In the ideal case, you can completely avoid computer processing and use an external recorder. The RODECaster Pro mentioned above would be a nearly ideal solution, though it’s not cheap. Of course, such a recording should be sufficiently “clean” from the start, but you can give yourself some slack and allow minor flaws. The time you save this way may well justify the imperfect result quality. As always, approach this with understanding and don’t take this recommendation as “say whatever however, the main thing is minimizing effort.”

distribution of this document

This text is distributed under a variant of the Creative Commons, Attribution-No Derivative Works license. You may use this text for any purpose (commercial and non-commercial), publish it on your websites, or distribute it in any other way, provided you credit the author and link to the original. Modification of formatting, splitting into chapters/parts is permitted as long as the original text remains unchanged.

If these terms don’t work for you, contact the author to obtain special permission.

This post was translated from the Russian original with AI assistance and reviewed by a human.