Jay Lake (jaylake) wrote,
Jay Lake

Lessons learned doing my six story challenge (long post)

As I mentioned here, with subsequent updates, I did a writing retreat this past weekend with a six story challenge. I wrote 26,200 words of first draft, with five finished stories. This was an interesting process, and it generated (or validated) certain insights for me about writing speed, productivity and career expectation.

Disclaimer: As many of you know, I have a day job in technical marketing. Therefore I have a tendency to try to reduce processes to numbers. I'm making a number of assumptions and SWAGs1 in this post. Take them with a huge grain of salt. I'm not making assertions about anything, even my own practice. Rather I'm just framing my thoughts about productivity.

Warning: very long post under cuts following

Voice, and what it means to shift gears so fast

I've posted on voice a couple of times recently, here and here to be specific. One aspect of setting myself this challenge was recognizing the need to shift voices. The process would have been different if I'd tried to write six stories in the same setting, or with the same character. I could have stayed in voice, in flow, and the net effect would have been more like writing a series of novel chapters.

As it was, my output varied considerably. In order, from Friday evening to Sunday evening, wrote:

TitleWord CountComments
Where the Water Meets the Sky2,400Near future utopian SF
C.V. of Pericles Chang800Object fiction
(experimental SF)
A Different Way into the Life5,600Contemporary urban fantasy
Crossing the Seven14,100Picaresque gonzo fantasy
The Leopard's Paw3,300Swords and sorcery
(inspired by Robert E. Howard)

There is something of an arc there, me not stripping my mental gears, though I did not notice it until just now. I started firmly in near-future SF, moved to contemporary urban fantasy, then crossed into more and more anaturalistic fiction. In other words, the degree of focus required in my thinking changed in a smooth progression.

At the same time, those are five distinct story types, setting and lengths. Shifts in length also substantially change the focus required from the writer brain.

I found a need for downtime of an hour or so moving from story to story. These aligned on various meal breaks and socialization on my part. (I took Saturday evening off from 4:30 on to go to a small dinner party with bibliothec, which certainly served to clear my head, as I did not get back to writing until Sunday morning.)

The basic point to me is that I can shift mental gears quickly, and produce work I feel is up to my standards. It's a valuable lesson, though not a surprising one.

There's another point here I need to acknowledge, and that's the issue of fast writing. I know a number of people are uncomfortable with the idea that something written so rapidly can reflect an author's best work. Every hour of classroom training we receive from grammar school to graduate school tends to emphasize the opposite: revise, polish, edit, improve.

I've said this before, and I'll say it again: for me, fast writing taps most deeply into the vein of my auctorial voice. I am not a brilliant writer. I am good (right now) at one thing. I don't shiny plots like Stephen King. I don't create deeply memorable characters like J.K. Rowling. I don't make the language my bitch like James Lee Burke. What I do well is voice. I'm decent at most of the rest of that stuff, and getting better, but right now I do voice. And voice comes, for me, with writing where my conscious mind, my inner editor, doesn't have time to step into the text, muck around with it, criticize my work.

If you're one of those people who reads me talking about writing fast and says, "it can't be that good if he wrote it that fast," well, I guess I disagree with you. Judge for yourself from my work, my career, my award and publication history.

If you're one of those people who reads me talking about writing fast and says, "he's a freak, that won't work for me," I have a suggestion. Try it. It certainly won't work for everyone, maybe not even for most people, but it's a neat trick if you can master it. You might surprise yourself with some good work.

Think about this, too: being able to write fast also makes a significant contribution to your ability to have a professional career. More on that point below.

How do we measure writing speed, and what does it signify?

Writing speed is a funny thing. As I see it, there's ramp up/ramp down speed, sustained speed and burst speed.

Ramp up/ramp down speed

Airliners fly around 600 mph. So why does a 600-mile flight take 90 minutes or more?

Imagine an airliner taxiing out at 10 mph, taking off around 140 mph, climbs to cruising altitude at increasing speed, then settling into cruise speed at 600 mph. It slows down to land, shedding speed on final approach, and touches down around 140 mph before taxiing in.

Writing is the same way, for me and probably for almost all other writers. No matter how mentally prepared you are, you don't sit down at the keyboard and blaze into full throttle on minute one. (Actually, this has happened to me a few times. It's a rather weird experience.)

I don't really have a method (or a need) to quantify ramp up/ramp down speed. Rather, it forms a sort of rounding-down factor when assessing overall writing speed. For example, when budgeting time for a project. See more on this below.

Sustained speed

Sustained speed is like the airliner moving at 600 mph. Once you're ramped up and into the story, and the mental prep, research and so forth are all squared away, this is how fast you work.

One of my long-term craft goals is to be able to write productively at my nominal typing speed. (FWIW, that would be about 3,600 words per hour.) I don't expect to ever achieve that level, for a variety of reasons, but I hope to asymptotically approach it over time.

In my own case, I am an extremely inefficient writer. I am probably adult ADHD -- never been clinically evaluated, but I usually max out the indicators those self-assessments you sometimes see. I cope with it just fine, largely by being very fast at most work tasks (writing or day job), so that my overall productivity essentially wipes out my efficiency penalty.

If you were to watch me write, you would see the keys flying for a couple of minutes. Then I'd jump into Google to check a date or an alternate spelling. Type another minute. Stare into space while I work out some scene blocking. Type two more minutes. Stop and check email, because if I do the same task too long I get very twitchy. (Why I stopped writing code, frankly.) Type another minute. Load Google News and see if anything exploded in the last twenty minutes. Repeat as necessary.

That's a work pattern that I suspect would be deadly for some writers, and disruptive for many. I simply don't work in a straight line.

The net impact is that my sustained writing speed is substantially less than my nominal typing speed. Experience has shown me that my sustained speed, when deep in a project, is about 2,500 words per hour.

Burst speed

This is when the juices are really flowing and I'm hitting it hard as I can. I rise above my sustained pace, set aside my ADHD-like ways, stop hearing when people talk to me, and type so fast my fingers hurt. My burst speed is north of 3,000 words per hour, approaching my nominal maximum. Burst speed isn't reliable, I don't count on it for anything, but it is an indicator of what level of productivity I can reach.

The difference between novels and short fiction, with respect to what has already been noted above

In the context of this discussion, I've also arrived at some conclusions about novels and short fiction. For purposes of simplification, I'm going to lump all short fiction together in one category and all novels in another. If I were doing a real business-grade analysis on this proposition, I'd break into six or seven categories, specifically:

CategoryWord Count
Flash1,000 words or less
Short1,000 - 7,500 words
Novelette7,500 - 17,500 words
Novella17,500 - 40,000 words
Short Novel40,000 - 75,000 words
Mid-Sized Novel75,000 - 150,000 words
Long NovelGreater than 150,000 words

(Note that these lengths are not arbitrary. There are significant differences in craft, technique, structure and plot sizing with those steps. There are naturally significant exceptions to each case as well.)

On to the simplified analysis.

Ramp up/ramp down speed is a significant factor in short fiction, meaning that it compromises a measurable portion of the effort in producing a story. For shorter work, the entire idea may emerge and be dealt with before achieving sustained speed. Even on a novelette or novella, that represents a decent chunk of the effort. Likewise, burst speed rarely enters into short fiction except in cases of extreme inspiration or works at the longer end of the range.

I stated above that my sustained speed is 2,500 words per hour. Looking at the amount of work produced and hours of effort involve in my six story challenge, I'd have to put my average rate of production for short fiction at about 1,500 words per hour, counting bathroom breaks and whatnot during the work session. This is net speed, a fourth way of thinking about writing speed.

What this means is that if I were budgeting time to write a 6,000 word story, I should allow about 4 working hours for the first draft. This does not account for research or other prep time, and it does not account for revision or marketing on the backside -- simply what it takes to get the first draft hammered out. But remember, the first draft is the pig iron of the writing process.

Novels, of course, are a different kettle of fish. A significantly smaller percentage of total time is consumed by ramp up/ramp down. There are also increased opportunities to work at my burst speed in those longer projects, have a 20,000 word day for example.

I know from recent experience with Other Me, Death of a Starship and Trial of Flowers that my average rate of production (i.e., net speed) for novels is about 2,000 words per hour. This means if I were budgeting time to write a 100,000 word novel, I should allow about 50 working hours for the first draft. Again, this does not account for research and other prep, nor does it account for revision or marketing.

See, this isn't just an academic exercise. Now I have a planning tool to help me evaluate deadlines. Obviously this only works for me because my writing behaviors are extremely consistent, but that's part of being a professional -- consistency.

Conclusion: I write novels faster than short fiction, on a words per hour throughput basis. I'd assumed that for a while, but I've never quantified it before right now.

The economics of a writing career, in a very rough cut

So now I'm going to map these numbers out to economics. Remember, these are very simplified cases. Assumptions include:

  • Every draft is successful, i.e., can be sold -- no time wasted

  • Marketing is a perfect effort, i.e., all work written is sold by me or my agent

  • No loss of writing speed is experienced (due to illness, stress, etc)

  • No adjustment for research time, since this is highly variable by project

Note these are not realistic assumptions, at least not for me. Rather, I don't have to build in fractional multipliers to account for them. (I could, remember, this kind of analysis is what I do in the day job, but the result, while significantly more accurate, would be annoyingly complex and rather less useful for the purposes of this discussion.) I'm mostly emphasizing this disclaimer because I don't want people to read this and then say "Jay said he can make $XX writing fiction." This is simplified, perfect-world stuff, intended to help me (and maybe you) think about the economics of a writing career.

I said above that I experience a net speed of 1,500 words per hour for short fiction and 2,000 words per hour for novels.

For short fiction I'm going to add a 50% time multiplier for revisions and marketing time, and no multiplier for downtime. In other words, I don't do extensive revisions on most short fiction, and the send-outs are a fairly trivial time investment on a per-story basis. Anent the downtime, I don't seem to need much between stories. A night's sleep is plenty. (This does not mean I could write short fiction continuously, day in and day out, eight hours a day. It simply means that when I do write short fiction I don't need to build in long breaks. Also, remember for discussion purposes I assumed "perfect marketing", which means stories sell in the first few sendouts.)

So for every hour I spend writing short fiction, I spend another 30 minutes revising and marketing. (Here's where the length assumptions go wonky, by the way. That's a ridiculous statement when applied to a 2,000 word story, but reasonable at 6,000 words, and bit ridiculous in the other direction at 20,000 words.)

For novels, I'm going to add a 100% time multiplier for revision and marketing time (though actual marketing load is low), and another 100% time multiplier for downtime. In other words, it takes me as long to revise a novel as it does to write it in the first place. I have also found I cannot simply slam from one novel to the next, and seem to require significant downtime between efforts. Taking my 50-hour first draft, that means another 50 hours of effort to reach revised draft, and 100 hours of down time from novel writing.

So for every hour I spend writing a novel, I spend another 3 hours revising and taking time off to reset my novel brain.

The above assumptions, while simplified, are based on my real life experience of the last year. They are not universals, though analogous assumptions should apply to any writer.

Let's take it back to words per hour.

I get 67% efficiency on my short fiction, following the formulation stated above. In other words, I spend 4 hours writing a 6,000 word short story in first draft, another 2 hours revising and marketing it, so I wind up at 1,000 words per hour as a paid working speed. (Yet another definition of speed. Follow that? It's the time to draft, plus the time to revise and market, which results in a paycheck given that I've assumed perfect marketing for discussion purposes.)

I get 25% efficiency on my novels, per the above, I other words, I spend 50 hours writing a 100,000 word novel, 50 hours revising it, and 100 hours letting the plot engine in my brain reset. That works out to 500 words per hours as a paid working speed.

Assume $0.05/word for short stories and $0.10/word for novels ($10,000 advance on 100,000 words). Again, remember this is "perfect marketing".

That means my 1,000 words per hour on short stories make me $50 per hour. My 500 words per hour on novels make me $50 per hour.

Weird, huh?

If I wrote full time (i.e., no day job, investing 200 hours per month in writing), at my current production parameters, in a perfect world I could write 12 novels a year, and make $10,000 per month doing it. Or I could write 8 novels a year, spend four months of the year writing 800,000 words of short fiction (not a typo, if you've followed the math), and still make $10,000 per month doing it.

So what does that all mean?

I'm not sure what it means. This has been a discovery process for me, working all this out. I can see some specific lessons -- things which should be obvious, but I now have a logical grounding for to support my intuitive grounding. Remember, I'm talking about my writing behaviors in a perfect-world context, so take this with a huge grain of salt.

Observation 1: Increasing my paid working speed will increase the economic health of my career.

Observation 2: I can do this by increasing my sustained writing speed, but then I already knew that.

Observation 3: Reducing the downtime requirement for novels would make a significant bump in my paid working speed. This would increase the value of novels with respect to short stories in terms of my writing income.

Observation 4: Novels have an open-ended upside on compensation. Advances scale proportionate to sales. (This applies to upside downside as scarlettina noted in comments.) This scaling upside is not true of short fiction. So as my career progresses, the calculation will weigh more and more heavily on the novel side.

I don't think any of these observations should surprise anyone. But it's interesting to me to see them validated through specific data points about my own career.

I haven't gone to the trouble of building a spreadsheet to run these numbers, but I suppose I could. If anyone wants me to, say so in comments and I'll try to find the time to build a calculator. Frankly, I think obsessing with productivity numbers is a real danger to most writers. I don't obsess with it myself (despite the apparent evidence of this post), rather, this is my professional training -- analyzing how the numbers that make up a business work. And writing, for all that it is an art, is also a business.

So why did I do this? Ultimately, to set my expectations. I now have a more precise idea how to budget working hours for developing first drafts. I have a reasonable idea how to budget calendar time for committing to book deadlines, as well as short story deadlines. Those in turn help me be more professional, which hopefully means my agent and my editors want to work with me more.

If you stuck with the post for this long, good on you. Tell me what you think -- did I miss anything important, did I blow some key assumptions. 'Nuff said for me.

1 SWAG = "Scientific Wild Ass Guess", an intuitive evaluation technique frequently used in estimating software projects in absence of hard data.
Tags: personal, process, writing
  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

← Ctrl ← Alt
Ctrl → Alt →
← Ctrl ← Alt
Ctrl → Alt →