[{"data":1,"prerenderedAt":30},["ShallowReactive",2],{"blog-body-blog-post-anatomy-en":3},{"content":4,"toc":5},"\u003Cp>Readers do not abandon articles because they are long. They abandon them at the exact point where nothing in the next paragraph promises to be worth the scroll. That moment is usually within the first two sections, and it is caused by structure rather than by subject matter.\u003C\u002Fp>\n\n\u003Cp>This is encouraging, because structure is the part you can fix without becoming a better writer. Every section of a post has one job: to earn the reader's attention for the section that follows. When you can name that job, you can check whether it is being done.\u003C\u002Fp>\n\n\u003Cp>Here is the anatomy of a post that gets read to the end, part by part.\u003C\u002Fp>\n\n\u003Ch2 id=\"why-readers-stop\">Why readers stop\u003C\u002Fh2>\n\n\u003Cp>The reason is usually not that the writing is bad. It is that the piece has stopped making promises.\u003C\u002Fp>\n\n\u003Cp>A reader's attention is maintained by a small, continuous debt: something was raised and has not yet been resolved. A question posed in the introduction. A claim you have not yet supported. A specific outcome you said you would show. As long as that debt exists, the reader keeps going, because abandoning it feels like leaving something unfinished.\u003C\u002Fp>\n\n\u003Cp>Articles that lose readers are ones that settle all their debts in the first three paragraphs. The introduction states the problem, summarises the solution, and lists the sections. Nothing is owed. Everything after that is optional reading, and optional reading does not happen.\u003C\u002Fp>\n\n\u003Ch2 id=\"the-opening-paragraph\">The opening paragraph\u003C\u002Fh2>\n\n\u003Cp>The first paragraph has to do three things quickly, in this order: confirm the reader is in the right place, name the specific problem, and imply that a resolution exists.\u003C\u002Fp>\n\n\u003Cp>Confirmation is a sentence, not a paragraph. If someone searched for a phrase and landed here, one line establishing that this is about that phrase is enough. Anything longer is a delay, and delays are where clicks get reversed.\u003C\u002Fp>\n\n\u003Cp>Naming the problem should be specific enough to be recognisable. \"Content marketing is hard\" is not recognisable; everyone has already agreed with it and moved on. \"You have twelve posts that each get forty views and you cannot tell which two deserve the effort\" describes a situation, and the reader who is in it knows immediately that the article is for them.\u003C\u002Fp>\n\n\u003Cp>Then leave the resolution unstated. Say that there is a way to decide, and move on. The gap between the problem and its answer is the thing that carries the reader through the next two thousand words.\u003C\u002Fp>\n\n\u003Ch2 id=\"section-headings-that-pull\">Section headings that pull\u003C\u002Fh2>\n\n\u003Cp>Headings are read far more often than the paragraphs under them. Many readers will scan the headings first and read only the sections that seem relevant. That means the headings are doing two jobs at once: labelling the section and selling it.\u003C\u002Fp>\n\n\u003Cp>A heading that labels only is a noun phrase: \"Formatting\", \"Common Issues\", \"Key Takeaways\". These tell the scanner nothing they could not have guessed, so they get skipped.\u003C\u002Fp>\n\n\u003Cp>A heading that pulls makes a claim or asks a question the reader wants resolved: \"Why the second draft is always worse\", \"Formatting that survives a copy-paste into email\", \"How to tell a thin section from a short one\".\u003C\u002Fp>\n\n\u003Cp>The test is whether the heading would make sense as a sentence in the article's argument. If it would, it is carrying weight. If it could be moved to a completely different article unchanged, it is a label.\u003C\u002Fp>\n\n\u003Ch2 id=\"paragraph-level-craft\">Paragraph-level craft\u003C\u002Fh2>\n\n\u003Cp>Paragraphs do the actual work of holding attention, and they fail in three predictable ways.\u003C\u002Fp>\n\n\u003Cul>\n  \u003Cli>\u003Cstrong>The long run-up.\u003C\u002Fstrong> Two or three sentences of framing before the point. Readers learn to skip these, and they skip the point along with them.\u003C\u002Fli>\n  \u003Cli>\u003Cstrong>The uniform block.\u003C\u002Fstrong> Four to six sentences of near-identical length and structure. Visually these read as walls, and the eye slides off them.\u003C\u002Fli>\n  \u003Cli>\u003Cstrong>The dead ending.\u003C\u002Fstrong> A paragraph that finishes without leaving anything open, so the next paragraph has to restart the reader's interest from zero.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\u003Cp>The repair for all three is the same: put the useful sentence first, keep the paragraph to two or four sentences, and end it in a way that makes the next one necessary rather than merely available.\u003C\u002Fp>\n\n\u003Ch2 id=\"the-example-that-earns-its-place\">The example that earns its place\u003C\u002Fh2>\n\n\u003Cp>Examples are the most commonly wasted element in an article, because a generic example is indistinguishable from an assertion.\u003C\u002Fp>\n\n\u003Cp>\"For example, a business might use this to improve its workflow\" is not an example. It restates the claim with the word \"for example\" attached. It teaches nothing and occupies a paragraph.\u003C\u002Fp>\n\n\u003Cp>An example earns its place when it contains a detail that could be checked or imagined: a number, a named situation, an obstacle, a before and after. The specificity is what makes it memorable, and memorability is what makes an article get shared or returned to.\u003C\u002Fp>\n\n\u003Cp>If you do not have a specific example, the honest move is to cut the sentence. A claim without an example is short. A claim with a fake example is a credibility problem.\u003C\u002Fp>\n\n\u003Ch2 id=\"transitions-that-do-work\">Transitions that do work\u003C\u002Fh2>\n\n\u003Cp>Transitions get treated as decoration and are usually written as filler: \"Now let's look at…\", \"Moving on…\". These signal that a section has ended without giving any reason to begin the next one.\u003C\u002Fp>\n\n\u003Cp>A working transition names the next thing and makes it sound necessary. \"That handles the diagnosis. The harder problem is deciding which of the twelve posts to fix first, because the obvious answer is usually wrong.\" Now the reader has a reason to continue that has nothing to do with politeness.\u003C\u002Fp>\n\n\u003Cp>The pattern is: close the current point, name the next one specifically, and add the reason it is not yet resolved. Three sentences, and the middle of the piece stops sagging.\u003C\u002Fp>\n\n\u003Ch2 id=\"the-ending\">The ending\u003C\u002Fh2>\n\n\u003Cp>An ending has two acceptable forms. It can answer the question the introduction raised, or it can tell the reader what to do with what they have just read.\u003C\u002Fp>\n\n\u003Cp>What it cannot be is a summary of the article. The reader has just read it. Repeating the main points signals that the piece had nothing further to offer, and it is the most common way a strong article ends on a weak note.\u003C\u002Fp>\n\n\u003Cp>If the introduction asked a question, the ending should answer it directly, in one or two sentences, without further hedging. If the piece was instructional, the ending should be the first concrete action, stated plainly enough to be started today.\u003C\u002Fp>\n\n\u003Ch2 id=\"a-checklist-for-a-finished-post\">A checklist for a finished post\u003C\u002Fh2>\n\n\u003Cp>Read the piece once, checking only these:\u003C\u002Fp>\n\n\u003Cul>\n  \u003Cli>Does the first paragraph confirm the topic and name a specific problem, without resolving it?\u003C\u002Fli>\n  \u003Cli>Would each heading make a reader want the section, or does it merely label it?\u003C\u002Fli>\n  \u003Cli>Does every paragraph open with its useful sentence?\u003C\u002Fli>\n  \u003Cli>Does every example contain a checkable detail?\u003C\u002Fli>\n  \u003Cli>Does each transition give a reason to continue?\u003C\u002Fli>\n  \u003Cli>Does the ending answer or instruct, rather than repeat?\u003C\u002Fli>\n\u003C\u002Ful>\n\n\u003Cp>Any \"no\" is a structural fix, and structural fixes are usually faster than rewriting prose — one moved sentence often repairs a paragraph that felt irreparable.\u003C\u002Fp>",[6,9,12,15,18,21,24,27],{"id":7,"text":8},"why-readers-stop","Why readers stop",{"id":10,"text":11},"the-opening-paragraph","The opening paragraph",{"id":13,"text":14},"section-headings-that-pull","Section headings that pull",{"id":16,"text":17},"paragraph-level-craft","Paragraph-level craft",{"id":19,"text":20},"the-example-that-earns-its-place","The example that earns its place",{"id":22,"text":23},"transitions-that-do-work","Transitions that do work",{"id":25,"text":26},"the-ending","The ending",{"id":28,"text":29},"a-checklist-for-a-finished-post","A checklist for a finished post",1791511468158]