How a jump link works
A jump link is an ordinary link with a fragment on the end of the address — the part after the hash. That fragment names an identifier somewhere in the page, and the browser scrolls to whichever element carries it. If nothing carries it, the link quietly does nothing, which is why broken jump links go unnoticed for years.
The fragment never reaches the server. The whole page is fetched as normal and the browser handles the scroll, which has one consequence people often get wrong: a fragment does not create a separate page. Search engines treat the address with and without it as the same URL, so you cannot rank a section on its own or report on it as a distinct page.
Jump links work from outside the page as well as inside it. A link in an email, an ad or another article can drop a reader straight onto the pricing section or the one question they asked about, and that is usually where they earn the most.
Why jump links matter
They shorten the distance between a click and an answer. Somebody who asked a specific question and is sent to the top of a long page has to hunt for their answer again; sent to the section itself, they read it immediately. On a phone that difference is the difference between reading and leaving.
Google does something similar without asking you. It can send a searcher to the relevant passage of a page rather than the top, and where a page is clearly divided with named sections it has more to work with. It also underpins accessibility: the skip link that lets a keyboard user bypass the navigation is a jump link, and it is often the first thing an accessibility review looks for.
Common mistakes with jump links
Sticky headers are the constant one. The browser scrolls the target to the top of the window, the fixed header sits over it, and the reader lands halfway into a section wondering what they missed. It is fixed in the stylesheet by reserving space above the target, not in the link.
Keyboard focus is the mistake nobody notices. By default a jump can move the view without moving focus, so the next tab press carries on from wherever the user was before, throwing them back up the page. Making the target focusable solves it. After that: identifiers that change when the heading is edited, so old links silently break; vague anchor text like “see below”, which tells nobody what they are jumping to; and using jump links to fake depth on a thin page, where the sections are one paragraph each.
How to act on it
Give the sections that people actually ask about stable, readable identifiers, and keep them stable even when the wording of the heading changes. Write the anchor text as the destination itself, so it reads as a section name rather than as an instruction — the same rule that applies to any table of contents.
Test on a phone with a sticky header in place, and test with the keyboard alone. Then use them deliberately in campaigns: a jump link from an email straight to the relevant part of a long page removes a step from the journey and costs nothing to set up.