Lesson 6 of 17
The Lighthouse · stone 6Intermediate6 minute read

Finishing

Completing drafts instead of restarting or abandoning them.

Reviewed by BlurbBio editorial team, 2026-11-11

The Lighthouse on its headland, with the keeper’s cottage and a writing desk in the window.
The Lighthouse · Writing process and habit, stone 6 of 17

Finishing is its own skill

Starting a story and finishing one are different skills, and it is possible to practise the first for years without practising the second. A writer who has begun six novels has had six goes at an opening. She has had none at a final third, at an ending, or at making a whole draft hold together.

None of these moments is evidence about the book, so none of them is a reason to drop it.

The science-fiction author Robert Heinlein once set out a short list of rules for writers, and one of them is blunt. In paraphrase: finish what you start. The reasoning is practical. Only a complete draft can be revised as a whole, and only reaching an ending teaches you how endings work.

There are a few moments when dropping a draft is most tempting. One is the middle, when the pleasure of the opening is spent and the end is not in sight. Another is the arrival of a new idea. Another is the discovery that the opening is wrong, which invites a restart from page one. Another is a long gap, after which the draft reads like someone else’s work.

None of these is evidence about the book. A new idea looks better than a half-written draft for a simple reason: the idea has not yet had to survive being written.

Make the end something you can check

Define finished. For a first draft, a useful definition is that every scene from first to last exists on the page in some form, with the gaps marked. That can be checked. “When it is good” cannot.

A first draft is finished when it can be checked, and “when it is good” cannot be.

Count what is left. List the chapters or scenes still unwritten. If you do not outline, estimate: about two-thirds through, perhaps twelve scenes to go. A guess is enough to start from, and you can correct it as you learn more.

Date the steps. Divide what is left by what your routine really produces, taking your ordinary pace and not your best week. Set a date for each group of chapters, and leave some weeks empty, because some weeks will be lost.

Park what would pull you off course. Keep one file for new book ideas and one list for changes to earlier chapters. Write each down in enough detail to pick up later, then go on drafting as if the change had already been made. Set a date to open both: the day after the draft is done.

There is a real cost. Pressing on means writing later chapters on top of early ones you know are flawed, and some of those pages will be rewritten. What you buy with them is the ending, and the ending is what tells you what the opening should have been.

Example

A dated plan to the last page

A horror novelist is seventeen chapters into a draft and feeling the pull of a new idea. This is the page she pinned above her desk.

The Drowned Parish: what’s lefthorror

Written 2 February.

Finished means: every scene to the last page exists, rough, with gaps marked.1 Not good. Done.

Written so far: chapters 1 to 17. Left: 18 to 26, which is nine chapters. I manage about one chapter every two weeks when nothing goes wrong.2

Chapters 18 to 20 (the flood): by 16 March.

Chapters 21 to 23 (the vestry, Delia’s confession): by 27 April.

Chapters 24 to 26 (the bell): by 8 June.

Then two spare weeks.3 Draft done by 22 June.

Parked until 23 June4: the opening is wrong (it should start at the well, not the funeral). Also the new idea about the lighthouse keeper’s sister, which has five lines in the ideas file. Neither gets opened before then.

  1. “Finished means: every scene to the last page exists, rough, with g…”The end is defined in a way she can check. Quality is ruled out as a test on purpose.
  2. “I manage about one chapter every two weeks when nothing goes wrong.”The dates are built on her ordinary pace, and she admits the condition attached to it.
  3. “Then two spare weeks.”Slack. Nine chapters will not all go to plan, and the final date already allows for that.
  4. “Parked until 23 June”The doubt about the opening and the new idea both have a place to wait and a date on which they will be looked at.

Before and after

A writer’s account: what to do with a better opening

The one faultThe writer acts on each new insight by restarting from page one.

Weaker versionOpen with the trial (first version)

At chapter fourteen I understood that the book should open with the trial, not the voyage. It was obvious, and I could not think how I had missed it. I went back to page one that night. The new opening took three weeks and it was better. By the time I reached chapter five again, everything after it needed changing to match, and somewhere in chapter eight I understood that the judge should really be a woman. So I went back to page one. I have a very good first hundred pages. I have had them for two years.

Showing the weaker version.

The insight is the same in both versions, and in both it is correct: the book should open with the trial. What differs is what the writer does in the next five minutes.

In the first version she goes back to page one. The new opening is better, and that is the trap. Rewriting forward from it, she has a second insight and goes back again. Each restart is justified, and together they keep her inside the first hundred pages for two years.

In the second she writes the insight down and drafts on as if it were already done. The second insight goes on the same list. When she reaches the end she has nine changes to make and a whole draft to make them in. She also finds that three of the nine are no longer needed. Until the ending existed she could not have known which three.

The exception

When to break it

Finishing is a habit worth building. It is not a duty owed to every draft.

Some drafts should be dropped, and the useful line is between deciding to stop and drifting away.

Some drafts should be dropped. The story may no longer interest you, or it was an exercise that has taught you what it had to teach, or your life has changed and it is not the book you want to write now. The useful line is between deciding to stop and drifting away. Write down that you have stopped, and why. If the same reason turns up on three projects, it is telling you something.

Some writers keep two projects going and finish both, moving across when one goes cold. If your record shows finished work, your method is sound.

A draft can also be set aside for good reasons: a paid deadline, a family emergency, research you cannot do yet. Leave a note saying where you stopped and what comes next, and a date to look at it again.

And a finished draft is not a finished book. Reaching the end is where revision starts.

Check yourself

Quick checkabout 1 min

The shiny new idea

A writer is two-thirds of the way through a draft when an idea for a different novel arrives, and it excites her more than the one she is writing. Which response protects the draft without losing the idea?

Key takeaway

Finishing is a separate skill from starting, and it is only practised by reaching the end. Say what finished means, date the steps at your real pace, and give new ideas and old doubts a place to wait. The ending will tell you what the rest of the book needs.

Practise

Exercises for this lesson

One is open to try now. The rest are free with an account.

Apply

A dated plan for your own draft

Write a one-page finishing plan for your current manuscript: what finished means, what is left, dated milestones at your real pace, and where new ideas and doubts will wait.

Use it on your own book

With a free account you can work through this on a passage of your own and keep the result.