Every piece of communication has a single angle it is actually about. The Golden Thread runs through every element and tests its right to be there. The paragraph that doesn't survive is usually the one you wrote for yourself rather than for the piece.
I was reviewing a comms strategy for a client last autumn, about two-thirds of the way through the document, when I found a paragraph that stopped me.
It was beautifully written. The thinking was sharp. The observation was the kind that makes you nod slowly and want to share it with someone. And it had nothing to do with what the document was about. Not subtly off-angle. The wrong essay entirely. Someone had been writing about one thing, drifted into another, and decided the quality of the writing was reason enough to keep it.
I recognised the pattern immediately, because I have done exactly this. Multiple times, in pieces I was proud of. The drift feels like enrichment while you're in it. You're expanding the argument, adding texture, showing the reader you've thought more broadly. Then you read it back and realise you've written two pieces that are sharing a document.
What the Angle actually is
Working through Compelling Communication, I found a framework that named something I'd been struggling to articulate. Every communication has an Angle: the single clear theme that the piece is actually about. Not the topic. Not the subject matter. The specific claim the piece is making about that subject.
"Leadership in tech" is a topic. "Technical founders who learn to delegate in year three outperform those who don't by a measurable margin" is an angle. The topic is a domain. The angle is a position.
Once you have the angle, you have what the book calls the Golden Thread: a test you run on every paragraph, every example, every aside. Does this serve the angle? If it doesn't, it doesn't belong in this piece. It might belong in another piece. It might be worth writing on its own. But it can't stay here, regardless of how well-written it is.
Every piece of communication has one angle. Everything that doesn't serve it weakens it.
This sounds like a constraint. It's the opposite. An angle gives you permission to cut without guilt, because the test is clear. The paragraph isn't failing because it's bad. It's failing because it belongs somewhere else. That's a much easier cut to make.
Kipling's completeness check
The other tool from Compelling Communication that I use alongside the Golden Thread is Kipling's Six Honest Serving Men. The poem most people know from childhood: what, why, when, how, where, who.
Used as a completeness check, it catches a different kind of problem. The Golden Thread test catches content that shouldn't be there. Kipling catches content that should be there but isn't. Once you've stripped the piece back to its angle, you run the six questions to make sure you haven't over-edited. Have you explained why this angle matters to this reader? Have you told them when and where it applies? Have you answered the how?
Most comms that feel thin aren't thin because the writing is weak. They're thin because one or two of Kipling's questions went unanswered. The reader is left with a position but no pathway. Or a pathway but no reason to take it.
Running both together takes about ten minutes on a first draft and saves hours of revision later. The Golden Thread first, to establish what belongs. Kipling second, to check nothing necessary got cut in the process.
The paragraph you least want to cut
There is a reliable pattern I've noticed across dozens of drafts, my own and other people's. The paragraph that fails the Golden Thread test most badly is almost always the one the writer is most attached to.
This makes sense when you think about it. The paragraphs that serve the angle write themselves and get forgotten. The ones that diverge are usually the product of a tangent the writer found interesting mid-draft, a moment of genuine enthusiasm for a connected idea. That enthusiasm shows. The writing is often better than the paragraphs around it. And the writer, consciously or not, finds reasons to justify it staying.
The paragraph you least want to cut is usually the one that doesn't belong. That's why you kept it.
The discipline is recognising what's happening. The divergence isn't evidence that the paragraph belongs. It's evidence that the paragraph triggered a different piece. Save it somewhere. Write that piece when you're done with this one. But don't let affection for a good paragraph compromise the piece it's in.
Strong argument in the first half, two illuminating digressions in the middle, a conclusion trying to tie all three threads together.
One angle, stated in a sentence. Every paragraph checked against it. The digressions saved in a separate document for later.
How to run it in practice
Write the angle at the top of the document before you start, in a single sentence. Not the topic. The specific claim you're making. If you can't write that sentence, you're not ready to write the piece.
When the draft is done, go through paragraph by paragraph. For each one, ask: if the angle is X, does this paragraph advance X? Not relate to X, not sit near X. Advance it.
Every paragraph that fails goes into a separate document. Not deleted. Saved. The resistance to cutting drops sharply when writers know the material isn't gone.
Then run Kipling. What have I claimed? Why should the reader care? How does it apply? When and where? Who does it affect? Who is telling them this? Fill any gaps. Don't add material that fails the thread just to answer a Kipling question. If the answer to "how" requires content that doesn't serve the angle, the angle needs revisiting.
I've been running this on everything for about eighteen months. The drafts that come back to me are shorter. The feedback on them is more specific. When a client says something feels off, I can usually trace it to a paragraph that survived the cut it shouldn't have, or a Kipling question that got half an answer.
The reader doesn't know the test exists. They just experience a piece that knows what it's about, says it, and stops.
