URL Structure
URL structure is the way the addresses on a site are organized: the path segments, their order, the words in them, and whether each page is reachable at exactly one address. A good structure lets a person read where a page sits from its address and lets a crawler treat each address as one page.
What an address is made of
RFC 3986 defines the parts: a scheme, an authority (the host), a path, an optional query and an optional fragment. Structure is a decision about the path and the query. Google’s URL structure guidance asks for readable words rather than identifiers, hyphens rather than underscores between words, and a hierarchy that is simple enough for a reader to follow. None of that is a ranking rule; it is what keeps an address legible when it is quoted on its own, which is how a citation shows it.
One page, one address
The failure that matters most is a page that answers at several addresses: with and without a trailing slash, in two letter cases, over http and https, with tracking parameters appended. A crawler sees several pages carrying one text and has to guess which is the original. Google’s documentation on duplicate URLs describes the consolidation signals, of which the canonical tag and a permanent redirect are the ones a site owner controls. An assistant that cites a page cites the address it fetched, so the address that gets quoted is whichever variant it happened to reach, and a redesign that changes paths without redirects turns every past citation into a dead link.
A common question
Do the words in a URL change whether an AI assistant cites the page?
Very little on their own. The assistant fetches a page because retrieval found it and quotes it because a passage on it answered the question; the words in the path did neither. Structure matters through stability and uniqueness: a page that keeps one address for years stays citable, and a page that answers at four addresses divides whatever signals it earns among them.
Related terms
Sources
Address syntax: RFC 3986. Structure and duplicate handling: Google’s URL structure guidance and its documentation on consolidating duplicate URLs. No audit figures on this page.