Banging Words Together

What's your process? I hear you ask. Technically, I don't hear you asking it, but since you've settled upon this blog, I assume you have some bare level of interest in the question. It's one of those questions that often interest readers (and new writers), because there is no single universal process for all stories and suitable for all writers. There are habits, there are superstitions, and there are tricks of the trade, but there are no algorithms by which one automates the process of banging words together. On occasion the author must venture forth and adventure off of the main road and into the literary wilderness.

Rather than dictate Universal Rules from On High, I'll tell you what my general tendencies are.

• I do a lot of planning . . . but not all of it. The hardest part about story planning and worldbuilding (which is a fertile topic on its own for another blog) is knowing when to stop. You can't always be certain what you're going to need before you've composed the story; this makes it almost inevitable that you're going to create more than you require — more background, more world, more plot, more characters, more scenes, more places. As I see it, you don't need much background in order to begin, and it's easier to spot the gaps as you go along if you haven't overbuilt. All I need is a sufficiently detailed overview in order to interpolate the necessary details. This habit stems from my many years of being GM (the game master or, as D&D calls it, the Dungeon Master) in tabletop role-playing games. When your players are unpredictable, over-planning is not your friend!* Thus, I find it much more valuable to be able to improvise plausible details using the broad strokes as a guideline.
• I plan so I know what the story's about. A writer friend told me many years ago that it isn't enough to know what happens. I need to know what the story is about on an emotional level. I build enough of a plan so I can see where the emotional connections lie. Otherwise I feel like I'm flying blind.

Woman on Painted Bridge isn't a story about a bisexual art gallery owner falling in love. It's about how we represent the truth. Titania conceals the truth about who she is by shifting, chameleon-like, to whatever label fits the social group. Avery conceals the truth about himself by retreating into a magical world he created in his paintings. By contrast, the character of Skye always says exactly what she's thinking, with no filter. Titania's business partner is honest and forthright, and her chosen activities — carpentry and surfing — are bound to the real world, with no nonsense.

The Unassassin (my current work in progress) isn't about a young woman and her cat who delve into the afterlife to find a new Emperor to put on the throne. It's about tradition versus innovation: how we stifle what could be when we cling too hard to what used to be. Khaizumai is the new generation, an 18-year-old woman who inherited her father's role as Imperial Unassassin, who does not see the point of following all the old traditions. Theizo, her father, is the man who insists on teaching her the proper way of doing things, based on ancient philosophies that have stood unchallenged for centuries.

• I keep a story bible. This means different things to different authors, but for me, it means a reference of all the details I don't want to keep in my head. This can cover a lot of ground. Generally, my planning goes into the story bible, but I also update the story bible as I go.
• Details about the world. When I built the Southlands in Seer of the Sands, I started by naming the countries A, B, C, D and E. Not very original, I know, but I had to start somewhere. I decided upon forms of government for each nation, named the important ruling figures, and established its capital. I decided how large each nation was, by way of comparison to some specific real-world region. The nation has to have a population, so I determined what its overall total population would be, its density, and the distribution of races — how many humans, elves, orcs, and so on. As I named each nation, I made sure to reflect its unique linguistic history in the mix of sounds I choose.
• Details about the characters. Although Scrivener has a place for tracking character details, I sometimes keep my notes in an Excel spreadsheet document rather than in a word processor so I can make use of multiple monitors, unlimited screen size and preformatted columns. Here I track each character's age, hair color and length, eye color, height, body type, and other story-specific details about them (name of horse; favorite phrases, sense of humor, likes and dislikes). If I'm writing along, I want an easy reference to which I can return when I need to know how I've described something that has appeared previously.
• Details about the character arcs, plot and story structure. I knew before I started writing the second chapter of Seer of the Sands that that the book would have 14 chapters, and I knew what would happen in each. I don't always do this, but when I do, it's useful to keep a record of that structure somewhere handy. It tells me that a certain character's most important moment is to be found in Chapter n, so I know I must work on escalating to that level in Chapter n – 1. If I think of a new scene, I can place it somewhere >n or <n depending on whether it escalates that moment or unwinds from it. 
• I read the story aloud to myself or someone else. This is the best way I know to do line editing and to clean up prose. When you read aloud to another, you're not permitted to skip words or skim over paragraphs that you're sick of looking at. There's no shortcut to reading aloud. It also gives you a sense of whether dialogue flows correctly, or whether there are missing scene transitions, stray punctuation marks, or any misspelled or overused words.
• I keep a glossary of names I've already used. I don't like to re-use names or over-emphasize names from a particular area of the alphabet. This is useful so I don't end up with several different orcs all named Dzgugh. It's happened.***
• I write exposition in the third person, then throw it away. Not actually delete it, mind you — I just set aside that exposition and let it serve to inform character dialogue. Exposition sounds better when a character says it, because then it's more than a series of dry, dull facts. It's a life experience; it's a trauma or a happy memory or a lesson somebody had to learn. Exposition given through the eyes of a character contains a point of view. Of course, before I can really write that expository dialogue, I first need to know what the truth is (or what will serve as the "truth" for story purposes).
• I'm always re-reading. You can't always catch problems on the first, second, or fifteenth pass. Keep reading your story, because you will find problems that you never found before.

* Overplanning in D&D means you could spend several days building an optional encounter in an old roadside ruin that lies on the direct road to the haunted wizard's tower**. Instead of investigating the tavern where the adventure begins and speaking to the old man who informs the players of the quest, the players instead start a bar brawl, burn down the tavern, slaughter the villagers, fight the night watch, and end up imprisoned in a desert town a thousand miles from the action. Instead of the game I had planned in the rain forest of a scenic sea coast, I'm forced to improvise a prison break in a desert town that I literally just invented sixty seconds ago.

** Which is either the tower of a haunted wizard, or the haunted tower of said wizard. Or, potentially, both. You never know when you start improvising.

*** No it hasn't. But it could happen.