In This Article
Writing a first book is technically easier and emotionally harder than ever. The tools that smooth the mechanical parts (Scrivener, Atticus, ProWritingAid, Sudowrite) have matured into a genuinely useful workflow, while the volume of competing manuscripts has grown enough that the bar for what counts as finished has climbed with it. The work the writer has to do is increasingly the part the tools cannot help with.
Below is a current path from a workable idea to a manuscript a literary agent or a small press editor would take seriously, with the time investment and the practical milestones at each stage.
TL;DR
The pick: Start with a one-page outline plus three sample chapters before you commit to a full manuscript. Eight weeks of structured work delivers most of the signal you need.
Runner-up: Pick Scrivener for long-form structural editing, Atticus for typesetting at the end, and a focused writing app (iA Writer, Ulysses) for drafting.
Skip if: Skip AI co-writing for fiction if you intend to query traditional publishers. Most major literary agencies require disclosure of AI assistance, and several explicitly will not represent AI-assisted fiction.
For a deeper reference, see Khan Academy’s official course catalog.
Phase 1: The idea, the outline, the proof-of-concept
A workable book idea needs three things: a clear protagonist or central question, a distinctive narrative engine (the why the reader keeps turning pages), and a place in the market that you can name to a stranger in one sentence. Spend two weeks on the one-page outline; the editing of that page is the most leverage you have over the next twelve months.
Before committing to a full manuscript, write three sample chapters (chapter one, the inciting incident or major argument, and a later pivot). If those three chapters do not read with the energy you imagined for the book, the full manuscript will not either, and you save yourself nine months of work.
Phase 2: Drafting infrastructure
Scrivener remains the most useful tool for structural editing of long manuscripts. Its outliner, corkboard, and split-document features handle the non-linear reality of book writing far better than Word or Google Docs. For pure drafting (fewer distractions), iA Writer or Ulysses pair well with Scrivener for the underlying project.
Backup discipline matters more than the tool choice. Sync to two cloud providers, version-control with Git or Scrivener’s built-in snapshots, and export a Word backup weekly. Lost manuscripts are usually the result of trusting a single sync rather than a tool failure.
Phase 3: First draft
Target a daily word count you can sustain. For working writers, that is usually 500 to 1500 words on a writing day; for full-time writers, 2000 to 4000. Sustain rate beats peak rate over the duration of a book; a writer who can hit 1000 words four days a week finishes faster than one who hits 3000 once a month.
Resist editing during the first draft. Note structural issues in the margins or a separate file; fix them in the second pass. Cycles of edit-as-you-go produce slower, more polished partial manuscripts; full drafts produce books.
Phase 4: Structural revision
The first revision is structural, not line-level. Read the full manuscript without editing, write a chapter-by-chapter beat sheet, and identify the three biggest structural problems. Fix those before touching prose. ProWritingAid, Grammarly, and Hemingway are tempting at this stage; resist them. They are line-edit tools, not structural editors.
Get one or two trusted readers (not family, ideally other writers or beta readers from a writing community) to read the post-structural-revision draft. Their feedback is the bridge from your editing to a stranger’s experience of the book.
Phase 5: Line editing and polish
Line editing comes last. ProWritingAid and Grammarly are useful here as flag-raisers, not as final authority. Sentence rhythm, word repetition, and POV consistency are the human-judgment parts; the tools cannot substitute for a careful read.
If budget allows, a freelance editor at this stage pays back materially. Look for editors who have worked at a publisher in your genre; their judgment about what an acquiring editor will care about is worth the 1500 to 4000 dollar fee.
Phase 6: Querying or self-publishing
For traditional publishing, the query process still works through literary agents. QueryTracker remains the best database; Manuscript Wishlist publishes individual agent preferences. Personalize each query, follow each agent’s submission guidelines exactly, and expect a 6 to 18 month timeline from first query to deal.
For self-publishing, Atticus handles typesetting for print-on-demand and ebook output. Amazon KDP, Apple Books, Kobo, and Google Play Books are the four major distribution channels. Plan marketing spend before you publish, not after; quiet self-published launches typically reach low double-digit sales.
The setup, step by step
- 1
Outline plus three sample chapters
Two to four weeks. The single highest-leverage early work.
- 2
Drafting infrastructure
One day. Scrivener plus a writing app plus cloud and Git backup.
- 3
First draft
Three to nine months at a sustainable daily rate. No editing during this phase.
- 4
Structural revision
Six to ten weeks. Beat sheet first, then fix the three biggest structural problems.
- 5
Line editing and polish
Six to ten weeks. AI grammar tools as flag-raisers, not authority. Optional freelance editor.
- 6
Query agents or self-publish
Traditional querying: 6 to 18 months. Self-publishing: 4 to 12 weeks to launch.
FAQ
How long does it take to write a first book?
Realistic full-time timeline is 9 to 18 months from concept to query-ready or self-publish-ready manuscript. Part-time writers should expect 18 to 36 months. Faster is possible but rare and usually shows in the finished product.
Should I use AI for any part of the process?
Brainstorming, outlining, beta-reader-style structural feedback, and grammar pass are all defensible. AI-generated prose in your manuscript is risky for traditional publishing and produces inconsistent quality for self-publishing.
Is it worth getting an agent for a first book?
For literary fiction, narrative nonfiction, and most genre fiction with breakout potential, yes. For tightly niche nonfiction (technical, regional, specialist), small-press direct submissions or self-publishing often work better.
What about ebook versus print versus audio?
All three matter. Plan to publish in all three formats simultaneously if self-publishing. Audio production (ACX, Findaway Voices) runs 250 to 800 dollars per finished hour for narrator and production.
Bottom line
Writing a manuscript rewards discipline over inspiration. The tools and the process are well-documented; the part that has not changed is showing up daily and trusting the work to compound. Pick your idea carefully, build the infrastructure once, hold the line on the first draft, and revise honestly. Most manuscripts that fail fail at the structural revision stage, not at the drafting stage, and most of those failures are recoverable with another six to ten weeks of disciplined work.
