Select Page

Ah, process. Some people live by it. But most writers I meet, especially amateur writers, hate process. I’m a software engineer by trade—and I approach writing very professionally—so process is a very important part of things for me. Here, we’ll explore my approach to pre-writing and word-count estimates.

I’m what NaNoWriMo calls a “plantser”—that is, a person who plans but also does a lot of work by the seat of their pants. I’ve adapted my process to reflect this approach, trying to strike the right balance between planning and giving myself freedom to improvise as I go.

Let’s get this out of the way first: everyone approaches this differently. My process will differ from yours, and part of your journey as a writer will be finding the process that works best for you.


Before I can outline, I have to start with an idea. This must include a setting[1], a character[2], and a mechanic[3]. (Note: I did not say “conflict.”) “Mechanic” often takes the form of “magic system” for me, but it can also be something else, like a Big Idea in science fiction. I call this my internal pitch.

The Prime Phoenix (alternatively called Forcefinder Phoenix) internal pitch:
A story about Megan[1], a beleaguered forcefinder mage[3] living in the Framework[2] of New San Francisco, who is the mother of Rylan [one of my Big Four characters in my epic fantasy series The Prime Saga].

The Prime Awakening internal pitch:
Trey[1], a human enslaved in the floating city of New San Francisco[2], discovers that elves have been here all along—and are the ones who put him there[3].

Once I have those things in place, I feel it out to see if I’m energized by this combination enough to write a book. This is a gut-check, and I only ever write books that pass this check.

The next step is to find the conflict. I didn’t necessarily find this conflict as part of my internal pitch, but it’s necessary before you have a story. One of the best tricks is to ask yourself this: who has the most to gain, and who has the most to lose by the mechanic idea? That often gives you your protagonist and antagonist and a conflict.

The final step—and this is crucial—is to find an ending. You don’t absolutely have to have this, but I recommend it at this stage. At the very least, you need your ending state—where your character is at the end. Your actual ending needs to have something “surprising yet inevitable” that uses the Character Opportunity (that’s another blog post) and builds on it to solve the problem. I don’t always know this specific thing that happens before I outline, but I usually try to.

Got all that?

Now you can outline.

The brainstorm

The next step for me is to brainstorm a lot of stuff. This is everything—the world, the characters, bits about the conflict, developing the magic, just anything that might go into the book (or might not). This is often structured loosely or not structured at all.

The dotted line

The next phase for me is the “dotted line,” which is a pre-outline. It’s more of a bird’s eye view of the story, where I lay out all the “mountain peaks.” The key here is identifying the big set pieces and the seven Plot Points, if you want to use that structure. When I first started out writing, I stuck slavishly to Seven Point Plot Structure. But now I feel it out more organically, since I am able to instinctively build plots that basically espouse the same concepts without feeling as formulaic.

The one thing structurally that I force myself to always keep to is the 75% point, around which you need both the Character Opportunity and the Darkest Night. Dan Wells explains those (not necessarily using those terms) in the link above.

Now it’s time to outline

The final phase of all this is the actual outline. This is really just an expansion of the dotted line—in fact, I usually start with that document and just add a lot of text to it.

Here’s where things can really vary between different authors. Some will break the outline into chapters and scenes, going as far as to dictate the beginning, middle, and end of each scene. I don’t do this. I don’t even break by chapter.

My outlines get broadly broken by viewpoint and setting, usually, though it depends on what’s going on in the story at the time. This is easiest to show rather than tell, so here are some examples from my most recent outline. One thing I’ll note is that over time I’ve come to outline less, preferring to invent more stuff on the day I write the scene instead. This is just something I’ve developed a feel for.

Here’s a first example, which is a bit on the more detailed side. You can see it took me 150 words to outline an estimated 2000 words. Here I’m calling out specific things that I know I need to include at this point—everything else will be made up on the spot when I write the scene. You’ll see character notes here, too, and this all occurs in a single viewpoint:

Beam and the Remnant of Lusvunub
2k words
– Walking through town, various bits of worldbuilding. Is happy. Ruthless.
– Preparations heightening.
– Show the bourbon.
– Show the fireblade. It reminds her of an old life. She takes it with her, wishing she’d made one for herself. She was thinking of an old friend.
– Preparations heightening more.
– Somehow include mention that dead souls in Lusvunub always stay, wandering forever in the city.
– End the scene with her coming up to the lines of Remnant soldiers moving through the gate. She watches them go, checking her equipment and supplies. Magona is there.
– They talk. Show Magona dying on-screen. A spider walks across her body after she dies. Beam shivers. (Keep this subtle! This is our canonical proof that Cariel does not die.)
– Beam steps through the gate into harsh sunlight. (Don’t forget to have her dazed for a moment due to the gate.)

Here’s how the book looks closer to the middle of the outline, when things are more frantic. I have 10 characters in one place here, and the outline does not say which of them get viewpoints. That’s on purpose—I’ll feel it out when I write it. 255 words for 4000 words:

Rylan, Elanil, Imra, Phoenix, Beam, Dill, Small, Allain, Erodar, Shot:
4k words
– Everyone goes into the Framework to check on things. Shot is still there, mustering his troops. The Crew is huge, now. A veritable army.
– They talk about the situation. Can they use the Crew to fight the Twins, somehow? Should they?
– Rylan’s radio perks up suddenly.
– The Twins have actually gotten hold of a transmitter. They’re scientists, after all. They say that they are hunting the man who imprisoned them. If he does not show himself, they will destroy everything around them until he does, up to and including all of Valaralda.
– Wait—one of the Twins grabs the mic. He feels something…a descendant? Someone related to the soul of the Imprisoner. (It’s Dillon!) He says he is going to leave his twin and go to the other planet—Earth—to kill this descendant or hold him hostage until the Imprisoner shows his face.
– Well, shit. Nobody here knows who that is, but they now know a Twin is coming to Earth. They muster the troops.
– Allain wants to make sure Pano Sylrantheas is okay.
– Beam wants to make sure Lusvunub is okay.
– Phoenix leads them to the huntroom, where they port down to the surface. They split up into two parties at this point. A bit of discussion that Rylan wants to stay in the Framework, but Elanil wants to see Pano Sylrantheas. She wins out, because Rylan wants to stay with her above all else. Kiss.

See how there is just enough detail here for me to capture the salient points that I need? Now here’s an example from toward the end of the book. At this point I used 472 words to outline 25,000 words! (Below is just part of that outline.) I’ll explain why later:

25k words combined
– The device is finally ready. Dill comes down, drawing his father further toward Fennas Elenathon.
– They activate the device.
– This severely stuns Orym. Now he will finally listen to someone talking.
– But he is still flying, for some reason. Dill is on a fallcar (like in Forcefinder Phoenix), following him.
– They talk as they fly.
– They end up at Lusvunub.
– Orym sees the bourbon. He stops. He loves bourbon.
– There is finally a chink in his armor.
– Dill talks to him, manages to get him to relent, to realize the error of his ways. Very emotional, pull out all the stops.
– Orym is going to be okay.
– They settle to the ground.
– We show Trey, Arra, everyone else arriving at this moment.

And here’s my final example. This is a battle sequence that’s intended to use 10,000 words. But I only put 103 words in the outline for this sequence:

Rylan, Elanil, Orym:
10k words
– (blocking, blocking!)
– Lots of interesting magic. This is the part of the book where we pull out all the large-scale stops. All the armies. Everything. This is the only part where the Department of Magical Research gets to do their thing.
– Use: The Army of Mages; the live rylak; the dead rylak; the Army of Souls; the Prime Mage Army; the Lusvunub Remnant; the Devout; the Eglaria and their shadowslaves
– This ultimately ends when Rylan and Elanil convince both sides to stop. [how do they convince the “good guys”?—via dialogue]

Okay. Here we can see me doing a few things. The “blocking, blocking!” bit is your first clue as to why I only used 103 words for a 10k sequence. I use the stage term “blocking” to refer to detailed actions—Trey flies north, encounters Arra, hits her with his staff, gets hit back, falls, casts about for magic, finds his magic, blasts her back, etc. I never do blocking in the outline stage. Instead, I do it just prior to writing the sequence. This keeps me fresh and up-to-date on what’s happening in the book specifically at that time, and it keeps me from getting bogged down in outlining. Blocking for me is almost writing—and I don’t want to do too much of it in advance.

The second thing you’ll see here is the “Use:” line. This is a reminder to myself that I have a whole bunch of different factions that I need to remember here. Without this line, I well may have forgotten one of them when I did the blocking.

The last new piece is the square bracketed statement: “how do they convince the good guys?” Sometimes my outlines include questions like this, where I haven’t quite thought it through just yet. I color-code these: red means “answer now, before writing.” Purple means “answer when you write the scene.” The one above is colored purple. I don’t sign off on an outline until all the red is gone.

Word count estimates

Now we come to the most subjective part of this whole exercise: word count estimates. Yes, I fully estimate out my entire book prior to writing any of it. And my track record is pretty good: The Prime Souls (original title, now broken into Pyramid and Arrival) and The Prime Phoenix, my last two books as of this writing, both came in with first drafts within 1,000 words of my estimate.

How do I do it? In a word: practice. I’ve written around a million words as of this writing, and when you write that much you get a good idea of how many words it takes you to do a certain thing. I can broadly break it down into categories, like this:

  • Small connector scene: 500 words
  • Small scene, not much happens, not a lot to say: 1000-1200 words
  • A more standard scene where some stuff happens and/or there is a decent amount of dialogue: 2000 words
  • There are grades to that: sometimes I need 3000 words for a scene

No single scene of mine almost ever gets beyond 3000. (If it does, I need to chop it down with a hook in the middle.) But most of the time, especially in longer books, I’m estimating “sequences,” not scenes. Those largely break down into 4k, 5k, and 10k. A 10k sequence is going to be a very long and complicated battle involving many moving parts.

I’m not always estimating what I think the scene will take—sometimes I’m estimating what the scene needs. There are many times in The Prime Mages outline where I outlined 2k words for a scene that I actually think will take 1k. But by estimating at 2k, I’m signalling my future self that I want to stretch the scene out, to pace it down, to take my time.

I’ve crunched the statistics, and I found that for my outlining style, generally 1,000 outline words equates to about 20,000 words in the book. In the case of The Prime Mages (original title, now split into Mages and Redemption), it’s a bit less—only 11,000 outline words for a 241k word book estimate. This is because fully 45,000 words of this book must utilize complicated blocking—none of which is written at this stage.

The bird’s eye view

With The Prime Mages, I pioneered a new technique for visualizing my outline. I took the broad setting/viewpoint groupings from the outline and distilled them into a list with titles and word count estimates, further broken by part. (There are four parts in each of The Prime Saga books.) Here’s what that looks like for the second part:

PART TWO Estimate
Moonbase 4,000
Lorelei and Elasha 1,000
Allain and Erodar 1,000
The Airon Sea 17,000
The Palace of Memory 3,000
Ilyrion 4,000
The United Sky Cities 4,000
Lusvunub 1,500
Pano Sylrantheas 5,000
The Framework 1,000
The Department of Magical Research 4,000
Lorelei and Elasha (2) 2,000
Turovoite Upper Decks 1,500
Quynn 1,000
The Battle of Lusvunub 10,000
Part Total 60,000
Book Total So Far 128,000

This immediately showed me that I am hopping between settings a lot in this book. I’m okay with this, because it’s sort of my goal. It also gives me a nice overview of where I am and for how long, letting me decide if I think the overall structure works in terms of character and setting.


Scrivener! That’s the only software I use for all this. Even when I tabulate counts for the bird’s eye view, I’m just doing that in a text document with custom tabs. This lets me stay within a single environment for everything. And while Excel is great, I prefer to keep everything in text format.

Some people prefer to outline in dedicated outlining software, such as OmniOutliner. But since I’m a software engineer, I find that I tend to think vertically in text. I prefer the freedom to format the outline any way I see fit, and not be beholden to whatever the software wants me to do.

In conclusion

This rather massive post still only barely scratches the surface of my pre-writing process. I call this process “story development,” because it encapsulates so many more things than just outlining. I’ve spent three years refining it, and it continues to evolve every time. One of the trends I’ve noticed this year is the tendency to outline a little bit less, deferring more invention to the day of writing. I think I gravitated toward this approach because when I outlined more than this, I ended up inevitably changing more of it as I went.

Your process will necessarily differ from mine. How do you outline? What types of work do you do before you start writing the actual words? How much does your outline match the finished product?