Chunking
Chunking is the practice of breaking a long document into smaller, coherent sections — each one able to stand on its own — so that a retrieval system can pull the single most relevant piece rather than an entire page.
How it works
Feeding an AI a whole 4,000-word page to answer one small question is wasteful and imprecise. Chunking splits that page into focused sections — ideally each answering one idea — so the system can grab just the part that matters.
Good chunking respects the structure of the content, keeping headings with their text and avoiding cutting sentences in half, so each chunk still makes sense on its own.
How SiteMind uses it
SiteMind chunks your content along its natural headings into right-sized, self-contained pieces, then lets you search, edit, or delete individual chunks — giving you precise control over what the assistant can say.
Related terms
Embeddings
Embeddings turn text into lists of numbers that capture its meaning, so a computer can tell which pieces of content are about the same thing.
Retrieval-Augmented Generation (RAG)
Retrieval-Augmented Generation (RAG) is a method where an AI looks up relevant information from a trusted source before answering, so its replies are grounded in facts instead of guesses.
Vector database
A vector database stores content as embeddings and can instantly find the pieces whose meaning is closest to a query.
Put this to work on your site
Free 3-day trial, no card required.