Start, sustain, and finish

How to write a book while working full time

August 16, 2026 · The StoryProp team

“I want to write a book, but I don't have time.” It is probably the most common sentence in writing, and it is almost never literally true. Four twenty-minute sessions and one ninety-minute session a week — a shape most full-time jobs can absorb — comes to roughly 1,400 words a week, about 73,000 words a year: a YA draft in 43 to 54 weeks, a commercial novel in 57 to 71. The hours exist — before work, at lunch, after the house goes quiet. What doesn't exist is continuity. People who finish books around full-time jobs are not less busy than you; they have solved a different problem than the one they thought they had.

Do the arithmetic properly, because it is more encouraging than it looks. The Manuscript Academy's 2026 word-count guide, sourced to agent Noelle Falcis Math of Transatlantic Literary Agency, puts commercial fiction at 80,000–100,000 words and fantasy at 90,000–120,000 — that is the finish line. For the rate, the best-documented case in English is Anthony Trollope, who timed himself with a watch on the desk and published the numbers in his posthumous “An Autobiography” of 1883: 250 words a quarter hour, three hours every morning from 5:30. The rate is not the same as the output — what he held himself to was a self-imposed weekly quota averaging 10,000 words, with recorded weeks running from 5,000 to 28,000 — and the whole of it was achieved around a full-time job at the Post Office. Halve his thousand words an hour for a modern amateur at a keyboard and an hour a day for a year still comes to more than 180,000 words. Two novels. So the hour is not the problem. The problem is that the hour on the calendar and the hour at the desk are not the same hour: the plan assumes sixty writing minutes; on a busy schedule, you will be lucky to get twenty.

One hour a day is really twenty minutes

Watch what actually happens at 9:15 on a Tuesday. You open the manuscript and reread to find your footing. You try to remember what you decided last week about the timeline — did the funeral happen before Mara came back to town, or after? You reread a little more, now with a critic's eye, and start doubting a scene you had no intention of touching. By the time your hands are moving, half the hour is gone, and some nights all of it.

This is the real enemy of writing a book while working full time: not word count, but context reload. A full-time novelist re-enters the book from a short distance, because they were inside it yesterday for hours. Even Trollope, at his desk before six in the morning after writing the previous morning, gave the first half hour of his three hours to rereading — a sixth of a professional's session spent on re-entry. You re-enter from a long way off, every single time, and that same half hour swallows most of what you have. The fix is not to write faster. It is to start faster — to arrange things so that minute one is a writing minute. Everything below serves that one goal.

End every session mid-motion

The oldest trick is still the best one: stop before you want to. Not at the end of the chapter, not even at the end of the scene — in the middle of the motion, sometimes in the middle of a sentence. Ending at a natural break feels responsible, but it hands your next session a cold start and an open question. Ending mid-motion hands it a running start and an order to follow.

Pair the unfinished sentence with a note to self, written in the ten seconds before you close the laptop. Not a summary — an instruction. “Daniel lies about the chipped cup here. Mara notices his hands, says nothing.” When you come back on Thursday, you don't decide anything; you obey the note. Deciding is the expensive part, and Tuesday-you already paid for it.

Write decisions down so returns are instant

Most of the re-entry tax goes to reconstructing decisions you already made once — the ordinary work of keeping a long story consistent, paid for twice. Is the story told in one summer or two? Does Daniel know about the will from the beginning? You settled these questions weeks ago, in your head, on a walk — which means you get to settle them again tonight, from scratch. An unrecorded decision is not a decision. It is a rehearsal for one.

So keep a decision log: a running file, one line per decision, dated. “June 12 — one summer, not two.” “June 30 — cut the sister; her function moves to the neighbor.” This is not an outline. An outline is a plan for what you intend; the log is a record of what you have settled, including the reversals. Reading it takes a minute and replaces most of the rereading you currently do just to feel oriented.

This habit matters double if you draft with an AI partner, and the fuller working method — direct in small steps, lock what works, revise by pointing — is its own subject. StoryProp meets this standard by building the log into the project itself: the conversation is permanent and ordered, and the decisions you approve are promoted to records — canon, characters, story shape, confirmed style rules — that the agent reads before it writes anything. What you agree to gets written down; what you don't stays a suggestion. Thursday's session opens with Tuesday's decisions already made.

Progress that survives interruption

The other quiet killer is choosing units of work that only pay out when finished. “Tonight I'll finish the chapter” is a plan that a toddler's bad dream can bankrupt. If the chapter needs ninety minutes and you are interrupted at forty, you have made progress you cannot feel — and months of progress you cannot feel reads, from the inside, exactly like failure.

Choose units that complete inside twenty minutes: one beat drafted, one continuity error fixed, one decision made and logged. A session that produces nothing but a logged decision is a successful session. If Tuesday's twenty minutes established that the book opens on the day Mara returns rather than the day of the funeral, that is worth more than a thousand words of the wrong opening — and it survives any interruption, because it was finished the moment you wrote it down.

Plan the week, not the day

Daily quotas are how a writing routine turns into a guilt machine. The famous one is NaNoWriMo's: 50,000 words between 1 and 30 November, an average of 1,667 words a day. Set that beside a working week and the mismatch is obvious — one NaNoWriMo day is more than the routine in this article produces in seven. Nor did the quota hold for the people who publicly committed to it: NaNoWriMo's own 2013 numbers record 42,221 validated winners out of 310,095 participants, 13.6 percent, with the average entrant finishing the month at roughly 11,350 words, about a fifth of the goal. A full life will always break a daily promise, so make a weekly one instead: a shape, not a streak. Most sessions will be short and opportunistic — twenty minutes at lunch, twenty after the kids are down — and they count wherever they fall.

Then protect one deeper session a week, sixty to ninety minutes, defended on the calendar like a meeting with someone important, because it is one. Use it for the work fragments cannot hold: drafting a full scene, reading the last few scenes in sequence to hear the rhythm, thinking about structure. The short sessions feed this one — they leave notes, log decisions, clear small debris — and it repays them with the momentum that makes the next fragment easy to enter.

If you want a writing routine that actually holds, anchor it to fixtures that already exist rather than to willpower: the first coffee, the commute, the quiet after bedtime. Routines pinned to a time of day fail when the day goes sideways; routines pinned to fixed events bend with it. And judge the week, not the day — four short sessions and one deep one is a very good writing week for a person with a job, whatever any single Tuesday looked like.

What to drop

Rereading from page one is the busy writer's most expensive habit. It feels like diligence; it is actually a search for confidence and context — two things the note-to-self and the decision log now hand you in under a minute. Adopt a hard rule: reread the last half page, no further. The opening chapters will still be there in revision, which is where rereading belongs.

Research spirals are the other one. Research is legitimate work, but it expands to fill any container, and it feels like writing while demanding none of writing's nerve. Mid-draft, a question gets a bracket — [CHECK: how long does probate actually take?] — and the draft keeps moving; the brackets get a scheduled slot of their own. StoryProp keeps findings in their own place, saved with their sources and the date they were gathered, separate from canon — so a fact you haven't verified can't quietly harden into something the book treats as settled.

Drop, too, the ritual polishing of chapter one. Returning to the opening again and again feels like craftsmanship, but it is usually re-entry anxiety wearing revision's clothes — the opening is simply the part you know best, so it is the safest place to stand. This holds for a novel, and just as much for a nonfiction book, where the temptation is to keep re-sanding the introduction instead of drafting the argument. Forward is the only direction a first draft understands.

None of this makes the hours longer. It makes them contiguous. End mid-motion, log the decision, pick a unit that finishes, shape the week, defend one deep session — and the twenty usable minutes stop being twenty minutes of warming up. They become the next twenty minutes of one long, slow, continuous session that happens to have a job in the middle of it. That is how books get written by people with full lives: not with more time, but with less of it lost. Fifty-two weeks of that is about 73,000 words — a YA novel outright, and most of the way to any other kind.

The tax on writing around a job is re-arrival — reconstructing, every evening, what you knew on Sunday. StoryProp is built to make the return instant: the project conversation is permanent, settled decisions live in records the agent reads before it drafts, and every pass is kept, so a twenty-minute session starts at minute one instead of minute fifteen. That is the difference between a year of fragments and a finished draft.

← All posts