Writing A Novel In 24-Hours: My One-Day NaNoWriMo Experiment

TL;DR: I wrote a 50,000 word novel in less than 24-hours using dictation.

“You can drive a car with your feet if you want to; it don’t mean it’s a good f*cking idea!”

– Chris Rock 

Disclaimer #1: Quantity is no substitute for quality… 

Disclaimer #2: Sometimes deadlines happen. Sometimes you’re in a rut and need a push. Sometimes you just want to do something insane.

Side note: Some people have confused transcribing with writing when I discuss this. When I talk about writing here, I mean actually creating the story on the spot, maybe from an outline, but not copying from some other form, ie, typing. 

I’ve written previously about speed-writing in my small guide of writing tips called How to Write a Novella in 24-Hours where I discussed writing shorter novellas in one day. Ever since I wrote that book, I’ve wondered if it was possible to write an entire *novel* in 24-hours. I’d heard about group efforts and I’d bet anything that high-output writers like Stephen King or Kevin J. Anderson probably did it at one point or another – but they have too much respect for their craft to diminish it like some kind of hot dog eating contest. I have no such compunctions. 

Speed writing fascinates me because I think it’s a great way to get over the psychology of “writer’s block” and also to make yourself aware of weaknesses in your writing process. I don’t need to write my books any faster, but I’m always eager to find out if I can make my writing more efficient so I can spend more time on the parts that really matter instead of reaching deadlines wishing I’d done a dozen things differently.

My previous record was approximately 28,000 words that went into a larger book (I’m not saying which one, but it was a finalist for a major literary award.) My record for a complete story told from start to finish in one day, including the outline, was 23,000 words. That book, Game Knight, currently holds a 4.2 star reader rating on Amazon – which is one of my lower ratings, but I’m proud of it, because that means the book doesn’t completely suck.

I’ve found that writing fast doesn’t always equate to poor writing, in fact I think the opposite is often the case. Some of my best writing has been when I’m in the zone and look up and realize that half the day has passed and the word count doubled. Bad books are often the product of writers struggling to get through a story that their readers will eventually have to struggle through to read. What writing fast definitely does is create horrific dumpster fire first drafts (at least in my case.)

Now that I’ve hopefully established that talking about trying to write really, really fast is more than just an effort to say “Hey! Look at me!” (Unrelated: Be sure to check out my Discovery Channel Shark Week episode Andrew Mayne: Ghost Diver…) I’d like to go into a few details about my latest experiment.

I’d talked with a few friends in the past about the logistics of writing 40,000 words (what most writers organizations consider to be the minimum length for a novel) in one 24-hour period. While that seemed technically possibly – my greatest marathon of 28,000 words also included 6 hours of sleep, it did seem like it was at the very borderline of what I could do (I’m a horrible typist.) But it was something I thought might be fun to try just to see where I break. I toyed with the idea of doing it during National Novel Writing Month, then found out they consider a novel 50,000 words or more – double the amount that I’d written for a standalone project in 24-hours.

I put the idea on the shelf for later in the future, possibly after I’d had a one-on-one with Mavis Beacon and felt like I type at the rate I could think. But then last week as I realized that NaNoWriMo was about to end and I’d have to wait another year, something began to stir in me. Maybe I should just throw my hat over the fence and give it a shot? Even though I knew I’d fall short, I was positive that falling short would mean that I’d at least surpassed my previous personal best. It would be worth doing for that reason alone – also for the next time I got into a pissing match with another equally insecure writer about what it means to really be “high-output”, I’d have some useless number to throw out there that could just as easily have been made up.

I started talking myself into the idea and then ran it by a few friends to see what they said (secretly hoping they’d talk me out of it.) Nope, Justin, Ken and Simon all said it sounded like a grand idea, and my buddy Peter J. Wacks and I had discussed doing this in great length. Throw your hat over the fence, they all said.

I decided to explore the feasibility of doing it at my current writing speed. The last time I’d considered doing this was when I thought 40,000 words would count for NaNoWriMo. Now I had to do some math and figure out if I had to write in the bathroom and carefully manage my sleep the day before so I could do a 24-hour marathon.

I opened up Scrivener and started free-writing a scene where I had a clear idea of the beginning middle and end. I stopped at 25 minutes and put the word count in a spreadsheet. 

When I calculated what my peak words per hour typing (creating the story as I went along) it was 2,839 words (which looked similar to what I’d achieved in the past.) This would mean that I would have to write non-stop for 18 hours. While that was borderline physiologically possible for me, I wasn’t sure if that was mentally possible. I still needed to spend a few minutes before each chapter figuring out where the story needed to go. I couldn’t always jump into things. I also knew that 2,839 words per hour rate would start to drop the longer I typed. My hands would tire at some point. My casual writing pace is about 1,800 words per hour (again, we’re talking about creating the story as we write, not transcribing it from something else. We’re writing, not typing.) 1,800 words sounded like a better guess at what I’d be achieving the longer I went, so I entered that into the spreadsheet. When I did the math it came out to this: 1,800 words x 24 hours: 43,200 words. While that’s technically a novel, that’s not a NaNoWriMo novel. I was back to where I was before: The realization that I couldn’t use my current method to write a 50,000 word novel in 24-hours. I needed to give up any hope of that fools gold of a goal.

Unless I changed up my methods.

Fun fact: There is no rule about how you write a novel. Some writers use dictation, others write in longhand. While I didn’t think I could type 50,000 words of an original story in 24-hours without the use of some kind of illicit or licit drug, maybe I could use writing for some parts and dictation for others? Maybe I could switch back and forth?

Although I have some experience with dictation, I even built an iOS app a few years ago that used to natural language processing to write dialog, I’d never used it extensively. I found that if I spoke for too long my throat would get sore. What if I could avoid that?

First I needed to find out if was worthwhile. How much faster could I speak than type? A few days before I’d ordered a refurbished Google Pixel 2 phone so I could try their new Recorder app that did real-time transcription. I pulled out the phone and used the app to dictate a new scene for 15 minutes. When I uploaded the transcription to Google Docs and checked the word count, I was shocked. It averaged out to over 9,000 words per hour. Assuming that I only did half that rate, not counting breaks, it would only take me a little over 10 hours to write the novel. But I was certain I could maintain a good pace if I had a clear idea what I was writing and didn’t get lost in the weeds.

The spreadsheet settled it. I wasn’t just going to try to break my previous record, I was going to make a serious run at trying to write a complete novel from conception to first draft in under 24-hours.

The first thing I did was go on Twitter and tell my friends that I was going to try to break my record on the last day of NaNoWriMo. I didn’t explicitly say I was going to write an entire novel, I just said I wanted to see how much writing I could get done in 24-hours. This was my face-saving explanation in case I burned out and couldn’t reach the 50,000th word.

After that I put up a few polls to figure out what I should write. For some stupid reason I decided that I not only had to write the book in 24-hours, I had to come up with the concept and write the outline in that time period as well. This was an arbitrary decision I made, but in part because a too detailed outline would feel like pre-writing in my mind and somehow diminish the whole “writing an entire novel in a day” aspect. I liked the idea of the novel coming completely to existence in just one day. It wouldn’t make it a better book, not by a long shot, but at least psychologically I’d know that I could pull something off like this with zero preparation.

While people on Twitter were making suggestions and voting on which topic and genre they liked most, I was reading up on how to keep from getting a sore throat from talking too much. 50,000 words is about six hours of non-stop speaking. I’d only gone that long once before on a crazy marathon live stream that left me with a raspy voice.

The answer was simple: Water, lots of water. Constantly drink water as you speak. Your voice goes hoarse because your vocal cords get too dry. Keep them lubricated and they keep working.

With that bit of knowledge my next step was to take the most upvoted idea and some of the suggestions and work them into an outline. I’d tried to game the system a bit by pitching “Goblins” as a subject. I’d been obsessed with them forever and was sure I could come up with a ton of ideas. I also tossed in “Vampire televangelists” because it sounded funny. I was positive people would choose goblins. No dice. Vampire televangelists won by a large margin. Fair enough, I decided, I could work with that.

I spent about an hour on my outline – about six days, twenty-three hours short of what I usually recommend for a novel, but it was enough to get me started. I had a clear idea of the beginning, middle and end once I figured out my protagonist. I won’t go into detail here about how I chose the structure or did my scene construction. But I tried to keep things simple. Three acts. Escalation of conflict and chapters that were written around scenes with clear conflicts and revelation that moved the plot.

I would have loved to have spent more time on the outline. I think that’s where books really live or die. Outlines are the DNA of a story. If there’s a flaw there you don’t see until too late, the book isn’t going to make it. However, for this experiment, the outline was there to just keep me writing. My goal wasn’t to write a great book, or even a good book, to be honest. It was to write a book with a clear plot and a beginning, middle and end. I just wanted to put 50,000 words down on virtual paper.

I want to stop here for a moment and point out a couple of helpful tips about writing using dictation.

1. Let the robots add the periods and commas Use a dictation program that automatically adds punctuation. Google Recorder does this on Android and Otter.ai – which has a generous free tier does this on iOS, Android and desktop. Having to say aloud “period”, “comma” or “question mark” is demeaning in 2019. Software that asks you do that is horribly written. You don’t need to do that anymore. Most note taking apps don’t and that’s why I create my own iOS app a few years ago that added punctuation when hit the talk button after each sentence. It wasn’t the best solution, but it was better than 5000 words without a period or comma. The generation of applications that work off machine learning models are much better at this.

2. The pause button is your friendStart the recorder when you have something to say, pause when you don’t. A lot of the pressure I felt during dictation was the long awkward period when I was gathering my thoughts for the next few sentences. Once my friend Peter J. Wacks explained to me the magic of hitting pause between moments, everything changed. You’re not trying to make a complete recording of the whole session. You just need the words – not the silence in between.

Outline in hand, it was time to start writing – in this case dictating. I decided to just start with dictation and switch to typing when my voice got too tired. I wrote the first 10,000 words in a little over an hour. I drank so much water my vocal chords hadn’t even begun to hurt, however my frequent trips to the bathroom were a side effect. Just to put things in context, normally for me a 10,000 word day is a really good day. To hit 10,000 words in such a short amount of time made the experiment a success right out of the gate. 
Here’s a quick look at what my process was:
1. Do a master outline.2. Add a little more detail on each chapter right before I started dictating.3. When I finished each chapter I saved the audio file and the text to Google Drive. Next I ran the text file through a little program I wrote that added quotes to what it thinks is dialog. (I based this on the wonderful javascript natural language processing library called “Compromise.”

An hour later I moved from my computer to another part of our home to dictate because I was feeling a bit self-conscious while my girlfriend was in the room. We share an open office area, which is normally fine when we’re both typing away on our computers, but when I’m haltingly enunciating my it feels very awkward to have another human in the room.

I hit 19,742 words about two hours later. The idea for the book didn’t even exist four hours before then. This was turning out much better than I had expected. I now knew the key was keeping my vocal chords wet and make sure I had a clear idea where I needed to go in each chapter. Occasionally I found myself starting and stopping over and over because I had the premise but not the story. For the most part, it was pretty fluid.

At 12:30 am, less than five hours after I started “writing”, I’d hit 26,749 words. That was a new record for a self-contained project and close to my best writing day ever on a larger book. But it was also an important milestone because I was more than halfway to the 50,000 word goal of NaNoWriMo. All it would take was another five hours of solid dictation and an ending and I’d have accomplished what I set out to do. Spoiler alert: There are no last minute twists.

At 4:56 am I reached 41,300 words. According to my calculation I had fourteen hours to do the remaining 9,000, so I decided to get some sleep. I woke up about four hours later, totally okay with the idea of going back to sleep for a few more hours, but couldn’t. My mind was racing about the story. So I got out of bed, went to my quiet place and finished a little after 1 pm.

If we count the moment I signed off Periscope at 5:15 pm on Friday evening and when I posted my final update at 1:44 pm the next day, that’s 21 hours from start to finish to write a 52,000 word book. I could have kept going, but the story ended where it needed to, so I didn’t see much point in trying to make it longer, although, I think it would be entirely possible to have hit a much more commercial word count of 80,000 if I spend the extra few hours and started with a much more detailed outline, instead of something I hastily put together.
 What were my biggest lesson learned? 

1. It’s absolutely possible to write a novel in a day using dictation.
2. Water, lots of water.
3. Write where you don’t feel self-conscious.
4. Make sure your outline is as granular as you can make it. You want to spend your time writing, not trying to think about what happens next.
5. Dictation is great for flow and speed, but man, the rough draft will be the roughest possible draft you can imagine.

Final thoughts…I learned more from this 24-hour experiment than I could have possibly imagined. What started off as an act of hubris (from the guy whose last public act of hubris was made into a Shark Week special this year) turned into an amazing learning experience. I didn’t just learn a great deal more about dictation, as I had to turn my outline into story, I saw some of my weaknesses in much greater detail and came away with a lot of insight into how to make my planning stage even more efficient, and hopefully write even better books in the future, regardless if I type them or speak them.

What about the book?The Reach, my temporary title, is a book conceived, outlined and written in 24 hours…which I think says where I should set my expectations. While I haven’t had a chance to work on the rough draft and see if it has any value beyond a practice exercise, I can confidently say that this book is probably one of the best novels ever written by a single author…in a day. It’s also the only one I know about.