0 words again today, but I got the novel off to editor #2! Now it's real.
I also finally got the opportunity to run the novel edits through my new chapter scoring engine. I’ll share some of my original assumptions and what the data actually told me, because it’s fascinating. From these assumptions, I could start designing some best practices for my writing sessions moving forward to reduce my edits further.
These assumptions are for me only. Maybe they’ll be helpful for you, but I’m not proscribing anything here. Just documenting my thoughts and findings out loud.
ASSUMPTION #1: The more sessions I take to write chapter, the more errors there will be, especially if those sessions have mixed writing methods or long lengths of time between them.
FINDING: True. The data bears out that chapters written in 2+ sessions tended to have more edits recommended by my editor, with 3 and 4 sessions driving the highest outliers.
BEST PRACTICE: I should (try to) avoid dragging chapters out into multiple sessions wherever possible. The common advice about not stopping in the middle of the chapter is definitely appropriate for me, and supported by data.
ASSUMPTION #2: The longer a chapter is, the more likely it is to have edits made by my editor.
FINDING: TRUE! Chapters higher than 2500 average words had the highest number of average total edits. Chapters at or lower than average of 1000 words had the lowest average number of edits. When you think about this, it makes sense. The more words you have on the page, the more chance you have for errors. The revelation for me is what the “magic” number is when I start seeing more errors overall. And it appears to be 2500.
BEST PRACTICE: If a chapter is greater than 2500 words, I will start looping through it TWICE before moving on.
ASSUMPTION #3: The better I feel when writing, the fewer errors I will create. My mood has an impact on the number of edits I receive.
FINDING: False! There was no real correlation. In fact, the chapters where I felt good drove the biggest outliers of edits, which could be an indicator that being in FLOW causes more errors! Two of the four biggest outliers were “flow“ chapters.
BEST PRACTICE: The existence of flow could presage more errors down the road. I’ll start flagging chapters with a flow indicator to evaluate if it makes sense to review them a little more, especially if there are other factors that could potentially drag the chapter score down.
ASSUMPTION #4: Some writing methods produce more errors than others.
FINDING: My assumption was that my laptop would be the cleanest writing method and dictation-multitasking would be the messiest. WRONG. The laptop drove the highest amount of errors (to be fair, I did write the most chapters on it though so it is somewhat overrepresented). My phone drove the lowest number of errors, but it was also underrepresented. Laptop typing does not allow me to catch hyphenation or spelling issues in real-time whereas dictation tends to catch them better, which is why I think the laptop caused my errors. You rarely have spelling issues when you dictate. Also, I can catch dictation errors more easily because they are somewhat predictable and I can catch them with automation, whereas laptop errors are more difficult to detect.
BEST PRACTICE: Laptop is not the cleanest writing method. Dictating on my bike is cleaner, believe it or not. It’s because I’m watching the screen actively while I’m speaking. When I’m dictating at my desk, I’m walking around and staring at the wall and not always looking at the screen. Therefore, I adjusted the scoring engine so that chapters I write on my bike will now score slightly more favorably. This is another reason to get on my bike and write while riding!
ASSUMPTION #5: A novel has fewer overall continuity/story errors than spelling/grammar errors.
FINDING: True. 88% of the edits were spelling and grammar-related, and 12% of the total edits were story edits. Only about 40% of chapters received story edits.
BEST PRACTICE: N/A
ASSUMPTION #6: Certain parts of a novel produce more errors than others (ie. The murky middle)
FINDING: FALSE! The total number of edits stayed relatively stable per chapter throughout the novel. Same was true with spelling and grammar edits. The murky middle is a writing problem, not an editing one.
BEST PRACTICE: I’ll forget about the murky middle once I write “the end.” It won’t matter moving forward.
ASSUMPTION #7: Because many of the errors my editor found were routine spelling and grammar errors and easily catchable by the engine in the future, I should use the engine after writing each chapter in order to eliminate edits PRIOR to self-editing.
FINDING: We’ll see. But I didn’t design the engine to use while writing. That’s a new revelation. Using it as I go could be a really smart strategy!
Anyway, I learned a lot from digging through the data from the first edit of this novel. I understand my manuscript on a deeper level.
It goes to show you that data is everywhere in your writing; you just have to look for it.
These findings will help me spend the right time in the right chapters moving forward. When I write Book 2, I should see a drastic reduction in edits, and that’s very exciting.
