You finish a draft and everything you meant to say is in it.
Every step of the argument. Every objection answered before anyone raises it.
And it reads heavy.
So you assume something is missing and add another paragraph.
Two different problems, one treatment
Unclear writing and over-explained writing feel identical from the inside. Both leave you with the sense that the piece hasn’t landed.
They need opposite treatments.
Unclear writing needs another sentence.
Over-explained writing needs one taken out.
The tell is the sentence immediately after your best line.
Read it and ask what it’s doing. Is it a new thought — or is it protecting the thought before it?
If it’s protecting, you didn’t write it for the reader.
You wrote it because you were afraid of being misunderstood.
Where this shows up in my own work
My strength is finding the deeper architecture of something.
My temptation is then explaining that architecture until nobody could possibly misunderstand it.
Those two run together. The second is what the first costs me if I’m not paying attention.
So I’ve written the standard down. One idea per piece. No sentence that defends the sentence before it. The last line lands without explaining itself. Enough marks to make the shape, and no more.
Writing the standard down turned out to be the easy part.
Applying it to a draft I’d just finished — that I couldn’t do.
The invoice draft
Last week I published a piece about the £500 invoice I felt awful about sending.
Here’s how the first draft opened:
I once sent an invoice for £500 and felt awful about it.
A colleague had asked me to review a recorded call and give her my perspective. We had agreed my consultancy rate. I listened twice, picked out the details I thought mattered, and recorded around twenty minutes of feedback.
Accurate. Complete. It starts with me, and asks the reader to work out why any of it matters to them.
I gave the draft and my written standard to Claude and asked it to apply the second to the first.
The opening came back as this:
You sit down to write and the first thing that arrives is something you already know.
You’ve said it in meetings. You’ve said it to clients so often the words have worn smooth.
So you skip past it and go looking for something better.
The reader is now inside the problem in three lines. My story arrives afterwards, as evidence.
Those three dismissals were already in my draft. They sat two-thirds of the way down, after I’d spent nine paragraphs explaining the mechanism that made them inevitable. The line about paying years to learn something was mine too, buried in the same stretch.
What changed was the order, and what was left out.
Fifty-six words came out of that article. Almost all of them were sentences defending other sentences.
What that is and isn’t
I don’t outsource my truth. I outsource my blind spots.
The experience was mine. The judgement was mine. The distinctions, the worldview, the standard being applied — all mine, written down over months.
What I couldn’t see was which of my own sentences were doing no work. I’d written them, so they all felt necessary.
Different output. Same signal.
This only works one way round. If the standard is vague, a machine will give you competence — smooth, publishable, and indistinguishable from anyone else’s. It fills the space you left empty.
The constraint has to be yours before the tool can apply it.
Which means the work is upstream of the writing. What do you actually believe about how your writing should behave? Not what you admire in other people. What you’d hold your own draft to at two in the morning.
Write that down and you have something to edit against.
Without it, you’ll keep adding paragraphs.
What editing turned out to be
Last week’s piece argued that experience earns explanation.
This is the other half.
The explanation still has to be cut by someone who knows what you were protecting when you wrote it.
Everything the reader needed was in my first draft.
So was everything I’d added to keep from being misunderstood.
Editing is telling those apart.
Read back your last piece and find the sentence directly after your best line. What is it doing there?
I’m developing Signal Expression — a 90-day programme for experienced professionals whose expertise has outgrown the professional identity it was earned in.
The work is about identifying the judgement that is genuinely yours, giving it language, and building a coherent body of thought people can recognise you for.
If that’s what you’ve been trying to build, join the Priority List.
The list closes on 1 October and I’ll be in touch personally when the founding cohort opens.