What is Content Chunking?
- Content Chunking
- Content chunking is the structuring of a page into self-contained sections that each answer one question completely, so any one of them can be retrieved and quoted without the rest of the page. Content chunking matters because retrieval systems split documents into passages before indexing them, and a chunk that leans on its surroundings reaches the model incomplete.
01 In practice
A well-formed chunk names its own subject in the first sentence, answers one question, and stops. Length matters less than independence: a short section that assumes the heading above it is worse than a longer one that restates what it is about.
The test is destructive and quick. Copy a section out of the page, remove the heading, and read it cold; where a pronoun points at something no longer present, or the subject has to be guessed, a retrieval system will hand the model exactly that broken text.
Chunking changes how a page is planned rather than only how it is edited. A page conceived as a sequence of answers, each with its own question, produces retrievable sections by construction, whereas an essay with headings cut into it afterwards usually carries its argument across the boundaries.
02 Related terms
The terms next to this one.
Technical
Passage Ranking
Retrieval that scores sections of a page, not the whole page.
Technical
Vector Embedding
Text turned into numbers, so meaning can be compared by distance.
Core
AI Citation
When an assistant attributes part of its answer to your page.
Core
Grounding
Tying a generated answer to sources it can be traced back to.
The work behind it
Rank on Google. Get cited by ChatGPT.
The sources assistants quote and the pages search still ranks, one engagement, both surfaces.