Many thanks to award-winning narrator and author Travis Baldree for writing this essay and giving me permission to re-publish it.
There is a prevailing sentiment that people who deviate from ‘best practices’ are exceptional in some way that is unattainable. That it’s magic, and out of reach for the hoi-polloi, or that beginners shouldn’t be exposed to such deviations from the norm, lest they make poor choices about their own process that might harm their career or inconvenience others.
I think this is fundamentally wrong, and that this broadly accepted ‘truism’ harms more people than most folks understand.
As I think should be self evident to any human taking advice on the internet in the year of our lord 2026 – I’m not saying anyone should do any of what I’m going to talk about later. It’s information. They can do what they want with it. Which is sort of the point of this.
Here’s the thing – I’ve had three careers. I’ve been successful in all of them. I am not a special snowflake. There’s enormous luck involved, but I think I have enough data over the decades to say that there are commonalities in people who succeed beyond the average.
One of the big ones is being willing to evaluate the ‘best practices’ and decide what works and what doesn’t, and to be willing to sacrifice some sacred cows. (This in itself is a skill that can be developed.)
When this happens, and anyone mentions the particulars, there is inevitably a negative response. “What if a beginner decided to do things this way?” Or, “well, THEY can do that because they’re exceptional – but…”
All of a sudden that information is devalued. There’s fear in discussing it. It’s been set aside as “not attainable” or “dangerous advice/information”.
So, let’s talk about these concerns.
Making mistakes is part of learning. Would anybody dispute that their journey to mastery involved making a lot of them? If someone is new to a career or an art, and they see what someone at the peak of their craft does, and attempt to do the same thing, and they screw up…
So what?
Happens all the time. That’s basically life for every kid looking up to an adult who is competent at something. It’s not even weird.
That’s VALUABLE information. Okay, that particular thing didn’t work for them. They have also definitely screwed up in countless other ways that had nothing to do with that attempt. Because they’re new. Of course they’re going to screw up.
Let people fuck up. You have to got to allow people to fuck up.
And what do you gain by suppressing this sort of discussion, anyway? In casting it as ‘dangerous’? Maybe they fuck up a little less?
What you lose, though, is having this information available and non-radioactive for the – possibly small – population where it would have been exactly what they needed. A path to finding their process. And moreover, a validation that it’s okay to try weird shit out and see if it works for you without social condemnation.
I think a lot of people are selling themselves short. “I can’t do that. I have to do established rule X, because I’m not special. That’s only for outliers.”
I am 100% positive that there are a lot of folks out there who are not achieving what they could if they could jettison that belief.
So I’m going to talk about my process as an example. I don’t like doing this in a public forum, because I think the kneejerk response is so ingrained that it will invite drama that I don’t want to court, but I guess I think it’s important enough that I’m going to do it anyway.
Here are things that I do that are “not best practices”. I assure you, they are not because I am lazy or belligerent or think that I’m special and ‘too good’ to do things the ‘normal’ way. And if we were talking about writing or software engineering, I would have a similar sort of list.
Prep: minimal. Maybe 30 minutes.
Read: I cold read. Every time.
(Note – cold reading is not the same as not prepping)
Production: self engineered.
So let me dig into why I do those things. Because there are sensible reasons!
Let’s start with prep.
To begin with, I believe that the level and type of prep is dependent on content. If you’re reading a science textbook, or historical fiction, or a fantasy novel, the prep requirements are fundamentally different. My process is shaped by the type of work I tend to do.
Secondly, I think a lot of our ‘sacred cows’ of prep have to do with the way production used to work.
If you are traveling to a studio, and there’s a producer and an engineer there waiting, prep is paramount – because stopping to check Forvo mid performance is wasting multiple people’s time. It’s not efficient, and it’s rude.
That’s not the norm anymore. I’d hazard that most books recorded are at home and undirected.
If you have to stop to check a pronunciation, and nobody else is lying fallow while you do, who cares? Nobody else’s time was wasted. You’ve just moved your research time to a different moment.
Efficiency doesn’t have to take into account multiple parties now (again, duals have their own needs) … and this allows you new opportunities
to be efficient.
I’m going to take the classic example of the accent that isn’t revealed until page 593.
If you didn’t know this going into a studio and read with the wrong accent? Catastrophe!
At home for a solo read? Well, you’re going back to re-record some lines. That sucks, and it wastes your time, but it doesn’t waste anybody else’s. (Before you bring up duals/duet, I’ll get to that)
Ok, so, take that and run out the math for yourself. How often does this happen? In my experience, not very.
Maybe 3/5 times in several hundred books for me? How many hours spent fixing such an issue in a year if you didn’t have foreknowledge? Not very many. Sucky hours, but not many of them.
How much time to deeply pre-read all of those books for the same year? A whole lot more.
Now, if I take my minimal prep time (which does not require an actual read of the book) and compare that time savings to the time lost in re-recording some lines…
It’s not even close. The amount of time I saved to get other things done is astronomical. From a pure efficiency standpoint it just makes good sense – for me.
Obviously I want to still minimize the chances of having to re record things (I value my time!), so this is what my prep involves:
I quickly scan the text by chapter looking for major characters. I check the book blurb. I identify them and list them. If a nationality or part of the country is listed, I try to find it. Dialogue is often instructive too (aye, ain’t, dropped g, etc) Who are the a/b/c characters? This takes very little time.
I search for keywords like ‘brogue’ and ‘accent’ and ‘drawl’ and note them.
I use spellcheck and find all the words Word doesn’t recognize. Great for fantasy and finding all the made up terms.
I send a note to the author or publisher with a list of my assumed pronunciations for made up words, phonetically, with the syllable emphasis capitalized, and ask them to approve or correct
That’s it.
A half hour, maybe. This gets shorter with sequels, too.
I don’t read the book ahead. I literally do not have the time. And doing so didn’t provide me a tangible benefit I could perceive. (If it does for you, super cool!)
Then I execute the cold read.
During performance if I hit a word I don’t know how to pronounce, I stop and research. (Merriam, Cambridge, Forvo, Youglish) If I have not caught a made up word and need to confirm, I drop a marker and send a question to the appropriate person and keep going. I record my best guess as part of the read. When I get an answer I’ll go back and pick that up, if needed, but more often than not I guessed it right. (This whole scenario is very rare.)
Whenever a character is encountered I immediately save a representative voice clip for reference and roll on.
(Note that in the rare cases where another narrator is involved – I only do a few duals – I will take the step of more extensively checking for character details and pre recording a voice for any that I am responsible for. We will have a dialogue. I will also pre record pronunciations for reference. I value other people’s time and do my best not to waste it. But that isn’t the norm for the books I am generally recording.)
Engineering:
I engineer my own work (oh no!), and I clean my audio on the fly. I use Adobe Audition in multitrack, and I auto-heal noises, nose farts, clicks and squeaks as they happen in spectral view, recording with headphones.
This is not hard to do, takes seconds, and is something that can be learned – if it is something that you want to do.
Once a read is done I master it with my stack, paste tops and tails, and send it to my proofer. My stack hasn’t changed in years. This is a button press.
If any sounds made it through the master (most are obliterated – pre mastering saves a ton of false positives) she lets me know and I fix them during pickups.
For pickups, I record, then master the pickup with a hotkey, and do sub-syllable or mid sentence insertions to minimize the impact of the pickup and make it as transparent as possible. Again, very learnable – sibilants and consonants provide the cleanest points for sub-syllable insertions. Practice makes this straightforward and I am really pleased at how undetectable I can make a pickup at this point.
I do not listen to the mastered audio! My proofer does. I use the same proofer for every single book. She gets a ton of consistent work from me.
As a result of all of this, my turnaround from script delivery to publishable audio is very, very fast. For a 10 hour book I can reliably have that shipped and live in 2 weeks.
I also don’t have an engineering cost beyond my own time (which was efficiently spent cleaning sounds the moment they happened, or addressing them from my pickup notes as part of pickup recording).
Over time, I have transitioned all my publishing work to self produced, even for Audible and PRH and Macmillan. It was gradual, but the benefits eventually became obvious. My audio is 100% consistent from book to book – and turnaround is very swift.
This process is not built around duals – if I was primarily a romance narrator I might work differently. The challenge that presents is primarily a mastering challenge – I’m not set up to engineer other people’s audio, which means we need to do a match between them and involve an engineer for their audio only. That works, but I think it’s clunky and I would probably investigate a way to streamline it if it were more prevalent for me. Right now the time benefit to doing that isn’t there, though.
I have a process that works for me, that is far outside the ‘norm’ – but is also based on experience and time and accumulated data.
Okay, so should somebody else do any of that?
No idea. Only they will know. But there’s almost certainly somebody out there whose particular niche or work or brain this fits – and I would like them to be able to try it if it hasn’t occurred to them to work this way.
If somebody tries it and it doesn’t work for them… great! They learned something. Maybe it even led them to something better that improved their process. I would be delighted if that were so.
Maybe somebody tries and it’s a complete disaster. Should I not talk about it to ‘save’ them from that disaster? Even though it should be obvious from everything I have said that it may not be for them? Even though they are a functioning adult who can make and refine their value judgements?
I don’t think so.
I think the loss in obscuring it from the people it might help far outweighs the loss/learning experience for those it doesn’t. Even if all they learned was ‘be more critical of advice on the Internet.’
I want more people to succeed far beyond the average. (Okay, I realize that if that happens, the math just means that the new average is now higher, but I think you know what I’m getting at here) I want people to be less dismissive of ‘outlier success’ as something that you cannot work toward, summarily discarding information that might evolve what our ‘best practices’ are, as though they are immutable and cannot change with time and circumstance. As if they really are the ‘best way’, when they almost certainly are not.
Best practices are the lowest common denominator. I think we should all be allowed to aim higher than that. And we should be able to fail and flail to get there.