Editing: Drafting vs Editing, the Cuts That Matter

Drafting and Editing Are Different Modes

Writing has two distinct modes, and trying to do both at once makes both worse.

• Drafting: producing text. Getting ideas out of your head onto the page. Generous, fast, imperfect
• Editing: improving text. Cutting, reshaping, sharpening. Critical, slow, specific

Drafting requires a suspension of judgment; editing requires judgment. The drafting voice says "keep going"; the editing voice says "cut this". If they're both in your head at once, the drafting voice can't get anything done.

The practical move: draft without editing, then edit separately. Different sessions if possible. Different mindsets certainly.

The Rough First Draft

Your first draft should be bad. Not aspirational: literally expected to be bad. The first draft's job is to get the thinking out of your head into visible form so you can work on it.

If your first draft feels polished, one of two things is happening:

1. You're editing while drafting, which slows both
2. The piece is short and simple enough not to need a rough stage

For anything more complex, rough-first-draft-then-edit is usually faster than trying to get it right as you go.

Writing rough

A few moves that help:

• Set a timer. 30 or 60 minutes of drafting. No stopping
• Write too much. Overshoot length; cut in edit
• Ignore transitions. Leave gaps; fill in edit
• Typo freely. Fix in edit
• Placeholder where needed. [STAT ABOUT X] or [FIND EXAMPLE] keeps you moving

At the end of a drafting session, you have material. Not a good document; material. That's progress.

The Edit Passes

A useful approach: edit in separate passes, each focused on one thing.

Pass 1    Structure: what's here, what's missing, what's in the wrong place
Pass 2    Clarity: are the sentences saying what they should
Pass 3    Compression: cut every word that doesn't earn its place
Pass 4    Surface: grammar, spelling, formatting, voice consistency

Each pass does one job. You can notice structural problems when you're looking for clarity, but fix only structural ones in pass 1. The discipline of single-focus passes catches more than trying to improve everything at once.

Pass 1: Structure

Read the full draft. Ask:

• Does this document lead with its main point?
• Does the sequence match how a reader would want to consume it?
• Are there sections that should be cut entirely?
• Are there sections missing that the reader needs?
• Are the headings descriptive of the content?

Structural changes are the most valuable and the most expensive. Making them in pass 1 means you won't polish sentences in sections you later cut.

Common structural problems

• Buried lead. The actual main point is three paragraphs in. Move it to the top
• Chronological default. The document is in the order you experienced things. Rearrange for the reader's needs
• Missing summary. No TL;DR; reader has to construct one from the body
• Disconnected sections. Each section is fine; the document as a whole doesn't flow

Pass 2: Clarity

Read sentence by sentence. Ask:

• Is this sentence saying what I meant?
• Would a reader understand it on a first pass?
• Is the subject of the sentence clear?
• Is the verb specific?
• Is there a better word for what I'm describing?

Specific clarity moves:

Replace abstract with concrete

Before:  "The team's performance has been somewhat inconsistent lately."
After:   "The team shipped two features in January but zero in February."

Name the agent

Before:  "A decision was made to postpone the launch."
After:   "The product leadership decided to postpone the launch."

Passive voice hides agency. Sometimes that's intentional; usually it's lazy.

Break up tangled sentences

If a sentence has multiple clauses joined with "and", "but", "which", and "while", it's probably doing too much. Split it.

Replace vague verbs

"Leverage" → use, apply
"Address" → fix, handle, resolve
"Enable" → let, allow
"Implement" → build, write, ship
"Execute" → run, deploy, do

Most vague verbs are replaceable. The specific version is usually better.

Pass 3: Compression

This is the pass where the document usually gets 20-40% shorter. Read paragraph by paragraph. For each one:

• Can any sentence be cut without losing meaning?
• Can any phrase be removed?
• Can any word be removed?

Specific compression targets

Wind-up phrases:

"In order to" → "to"
"It is worth noting that" → just write the thing
"As a matter of fact" → cut
"For all intents and purposes" → cut

Hedges:

"might possibly" → "might"
"somewhat" → often cut
"rather" → often cut
"just" → often cut (as in "I just wanted to...")

Redundant pairs:

"first and foremost" → pick one
"each and every" → pick one
"basic fundamentals" → pick one
"final outcome" → "outcome"

Throat-clearing:

"I think that..." → just make the claim
"In my opinion..." → just make the claim
"It should be noted that..." → state the thing

The "cut by 10%" test

A discipline Stephen King recommends: cut your draft by 10% in revision. If you drafted 2000 words, target 1800.

In practice, most drafts can safely lose 20%. Sometimes more. A trained compressor cuts without losing meaning; the document gets faster to read and more impactful.

Restraint on compression

A caveat: ruthless compression can strip voice. A document can become terse to the point of cold. Clarity is the goal; brevity is a servant, not a master.

If a sentence exists for rhythm, pacing, or warmth, it can stay even if it's not strictly necessary. Skilled editors know which "extra" words are signal (tone, voice) and which are noise (filler).

Pass 4: Surface

The final pass handles:

• Spelling
• Grammar
• Punctuation
• Formatting consistency (heading styles, code blocks, link formatting)
• Voice consistency (tone shifts within the document)
• Broken links
• Numbers that don't add up
• Typos

This pass is mechanical. Useful to do it cold, or with some distance from the content. Tools (spell checkers, grammar checkers, linters) handle some of this. The rest is careful re-reading.

The Cold Read

Between drafting and serious editing, set the document aside. A day if possible. An hour at minimum. Longer for high-stakes documents.

Returning cold, you read more like a reader and less like the author. You notice things: the sentence that sounded good but doesn't make sense, the paragraph that was clear to you because you lived through writing it.

Professional writers use this. Amateurs sometimes skip it; they pay for the skip in missed problems.

Read Aloud

A strong editing test: read your draft aloud. Your ear catches things your eye doesn't.

• Sentences that are too long (you run out of breath)
• Unnatural rhythm (the sentence doesn't land)
• Repetitive word choice (you hear the same word three times)
• Unclear pronouns (you can't tell who "they" refers to when spoken)

Not every edit needs this; for important documents, it's worth the five minutes.

Editing Other People's Writing

When you're editing someone else's work:

Preserve their voice

Don't rewrite their document into yours. Keep their idioms, their sentence patterns, their perspective. Edit for clarity, not to make it sound like you'd have written it.

Comment rather than rewrite

Suggest: "consider starting with the TL;DR here" rather than just rewriting the intro. The author learns more; the revision stays theirs.

Prioritise

Not every problem is worth flagging. Note the top 3-5 issues; smaller issues can be noted or fixed silently. A review that flags 50 things is overwhelming and usually doesn't get fully applied.

Be specific

"This section is unclear" is less useful than "I'm not sure whether you're recommending X or Y; which is it?". Specific feedback gives the author something to act on.

Be kind about priorities

You're reviewing their work. They've invested in it. Leading with "this is great; I have a few suggestions" is not false flattery if followed by specific feedback. Lead-with-praise is partly tactical and partly just human.

When to Stop Editing

A real question: how do you know when to stop?

Signs it's time:

• The changes you're making are small and stylistic, not substantive
• You've stopped finding clear improvements
• You're rewriting things that were fine
• You've been editing for longer than you drafted

At that point, ship. A document that's been over-edited often loses life; a document shipped at 90% quality usually works as well as one shipped at 95% and costs half as much time.

Perfection is expensive and usually imaginary. Good-enough-at-this-moment is fine.

Editing as a Long-Term Skill

Editing gets meaningfully better with practice over years. Specifically:

• You notice problems you didn't before
• You cut faster, with less hesitation
• You develop a sense for which sentences carry weight and which don't
• You spot your own patterns of over-explanation, padding, or fudging

Most of this comes from editing lots of your own work, reviewing other people's, and occasionally reading good editors' work on your text. There's no shortcut; there's just volume over time.

The Revision Loop in Writing Culture

A closing note on how teams relate to editing:

Good writing cultures have editing built in. Reviewers don't just rubber-stamp; they actually read and suggest. Authors expect and welcome edit feedback. Nobody takes revision requests as attacks on their ego.

Bad writing cultures dress up first drafts and ship them. Everyone pretends the document is good because no one has the energy to read it carefully. The output is readable but not good; nothing's ever really revised.

If you're in a bad writing culture, you can work on your own revision practice anyway. Over time, you'll produce noticeably better work than your peers, and you'll be in demand to review theirs.

Common Pitfalls

"I don't have time to edit." You don't have time to send unedited work that causes misunderstandings and rework. Editing trades upfront time for downstream time

"I'll edit as I go." Usually slower than drafting-then-editing. Your drafting voice and editing voice fight each other

"The first draft was fine." Sometimes. Usually, rereading after a day reveals it wasn't. The first-draft-as-final move is a sign of not using the cold-read pass

"Editing is mechanical." Some is (spelling, grammar). Most isn't. Deciding what to cut, what to reshape, what to add requires real judgment

"If I keep editing, it'll get better." Up to a point. Past a point, editing is fiddling; additional passes subtract more than they add

Next Steps

Continue to 09-naming-and-framing.md for the compressed interfaces that do outsized work.