Finishing a draft creates a peculiar temptation. You can finally see the whole manuscript, and almost every page contains something you could improve. A clumsy sentence catches your eye. A scene feels too long. A character disappears for eighty pages. The opening now seems to promise a rather different book from the one you eventually wrote.
If you begin correcting whatever you notice first, revision can become a long series of local repairs. Some will be useful. Others may be wasted when a later structural decision removes the passage altogether.
Before rewriting, find out what the manuscript actually needs.
Read the whole thing first
Give yourself enough distance from the draft to read it with some curiosity. Then read from beginning to end without trying to repair every problem as you encounter it.
Make brief notes. Record where your attention drops, where you become confused, where the pace changes unexpectedly, where a character’s behaviour stops convincing you, and where the manuscript suddenly feels alive. Keep moving.
This first read is about patterns. A weak paragraph is a paragraph. A problem that appears in twelve chapters is part of the manuscript’s structure.
Describe the book you have written
Set the original plan aside for a moment and describe the manuscript that now exists. What happens? Whose story is it? What changes between the beginning and the end? What question or pressure holds the material together?
This can reveal a gap between intention and execution. Perhaps the supposed protagonist has become less important than another character. A memoir planned around one period of life may have developed a second centre. A non-fiction book may contain two strong arguments competing for the same space.
Work with the evidence on the page. Revision becomes much clearer once you know which version of the project you are developing.
Find the problems with the widest reach
Some changes affect a sentence. Others affect fifty pages. Start with the problems that touch the greatest amount of material.
If the protagonist’s central motivation is unclear, dialogue, pacing and individual scenes may all improve once that motivation is repaired. If the chronology is confusing, polishing transitions before you settle the order creates extra work. If two chapters perform the same job, deciding which one belongs may remove pages you would otherwise spend hours editing.
Make a short list of the three problems whose correction would improve the greatest part of the manuscript. These are your first revision priorities.
Follow cause and consequence
In narrative work, look at the chain of events. What causes each significant development? What changes because it happened?
A manuscript can contain plenty of incident while still feeling static if events do not alter later choices. A revelation should affect what someone knows, wants or does. An argument should leave a residue. A decision should narrow or change what becomes possible next.
Where the chain breaks, investigate. You may need a stronger cause, a clearer consequence, or a different event altogether.
Look at character pressure
For each major character, identify what they want during the main movement of the story and what makes that difficult. Then look at how those pressures change.
Characters can appear inconsistent when the manuscript has failed to show the step between one state and another. A reconciliation arrives before enough has happened to make it credible. Fear disappears because the plot needs the character to act. A secondary character creates a problem and then seems to forget it.
Mark those missing steps. They are often more important at this stage than refinements to voice or dialogue.
Check what the reader knows, and when
Information has timing. A fact revealed too early can drain tension from later scenes. The same fact revealed too late may make earlier behaviour incomprehensible.
Track the important pieces of knowledge in the manuscript: secrets, motives, relationships, history, evidence, promises and misunderstandings. Note when the reader learns each one and when the characters learn it.
This is particularly useful when a draft feels confusing in some places and over-explained in others. The underlying problem may be the distribution of information.
Find repetition
First drafts often contain more than one attempt at the same thing. You may explain a relationship in narration and then demonstrate it in the following scene. Two conversations may reach the same conclusion. Three examples may support a point that only needs one.
Look for repeated information, repeated emotional beats and repeated functions. Decide which version carries the most weight, then see what can be cut, combined or reassigned.
Notice where the manuscript is strongest
Diagnosis should include what is working. Mark the scenes, passages or sections where the writing has energy and authority. Ask what they have in common.
Perhaps the viewpoint is closer. Perhaps the characters have something immediate to lose. Perhaps the prose becomes simpler when the subject is emotionally difficult. Perhaps your strongest non-fiction chapters are organised around concrete cases rather than general explanation.
Those successful passages can teach you something about the rest of the manuscript.
Build a revision order
Once you have diagnosed the draft, put the work in an order that prevents you from repeatedly undoing your own edits.
Deal with large structural decisions first: the book’s focus, missing or unnecessary sections, chronology, viewpoint, major character arcs and the order of material. Then move into scene and chapter work. Sentence-level editing and proofreading belong later, when the material itself is reasonably stable.
Your manuscript may need several passes. Give each pass a clear purpose. One might repair structure, another track a character, another tighten explanation, and another concentrate on prose.
Write a short diagnosis before you begin
Summarise the state of the manuscript in a paragraph. Name its strongest feature, the main problem, and the two or three changes that matter most. Then write down the first revision task.
Keep this note nearby. New problems will appear as you work, and some of your original judgements will change. The diagnosis gives you a starting order while leaving room for the manuscript to teach you more during revision.
A complete draft contains far more information than the plan you began with. Read that information carefully before you start rebuilding. The time spent diagnosing the manuscript can save many hours of beautifully editing pages that were never going to survive.