Answer Block

What makes an answer block extractable rather than just short?

Standing alone. An introduction assumes the reader has the title above it and the article below it; an answer block assumes neither, because a retrieval system lifts passages and discards their surroundings. In practice that means naming the subject rather than referring to it, stating the answer before the context, and containing no sentence that depends on a preceding one.

The test is mechanical and takes ten seconds. Copy the block into an empty document. If it still answers the question and still says whose answer it is, it works. If it opens with “this” or “it” or assumes the heading, it does not. The most common failure is not vagueness, it is context borrowed from the title.

A block reading “There are four of them, each with a different job” is perfectly clear on the page and meaningless anywhere else, and it is the shape most introductions naturally take. The second most common is a first sentence that promises rather than answers. “This guide explains how AI crawlers work” is a description of the page; “an AI crawler is an automated fetcher run by an AI company” is the answer, and only one of them is worth extracting.

How long should an answer block be?

Short enough to quote whole, which in practice means a few sentences rather than a paragraph of build-up. The supporting evidence for brevity is indirect but consistent: Ahrefs’ analysis of 560,346 AI Overviews found the average cited page at 1,282 words with 53.4% of cited pages under 1,000, and a correlation between word count and citation position of 0.04.

The glossary you are reading uses 40 to 60 words as a hard field constraint, enforced at save rather than as a convention, because a rule nobody can skip is worth more than a guideline everybody agrees with.

Is “answer block” a real standard?

No. It is an industry coinage with no specification, no schema type and no engine that recognises it by name, and the same shape travels under several labels including answer-first, the summary block and BLUF. Nothing is marked up as an answer block and no engine looks for one.

What is real is the position effect underneath it, which is why the technique survives its own vocabulary. Being early in a document and complete in a passage is what gets a page quoted, and the name for it does not matter at all.

Related terms

• BLUF (Bottom Line Up Front)BLUF, or bottom line up front, is the convention of stating the conclusion in the opening sentence and the supporting reasoning afterwards.
• Atomic ContentAtomic content is the practice of writing in self-contained units, each covering one idea completely enough to be understood alone.
• ChunkingChunking is the step that splits a page into passages so a retrieval system can select one without the rest.
• Information GainInformation gain is how much new information a document adds beyond what a reader has already seen elsewhere.