An HTML sitemap is not just a list of URLs organized by category. On a content-oriented site like Mes Rêves, the sitemap page displays the entire indexed structure, allowing for the identification of depth levels, orphan pages, and thematic groupings without the need for an external crawler.
Difference between XML sitemap and HTML sitemap on Mes Rêves
The XML sitemap is intended for indexing bots. It conveys metadata (update frequency, relative priority) that search engines use to plan their crawl. The HTML sitemap, on the other hand, is designed for human visitors: it displays page titles organized according to the actual hierarchy of the site.
On Mes Rêves, the HTML sitemap reflects the navigation structure as it is published, not what a technical file declares to the engines. A discrepancy between the two often indicates pages that have been created but are never linked to the main menu, or empty categories still referenced in the XML.
We recommend comparing the two versions to spot these inconsistencies. Consulting the structure of the Mes Rêves site in HTML provides an immediate overview of the actual linking, whereas the XML only delivers a flat inventory of addresses.
Multi-access navigation: why a single path is not enough
The W3C recommends combining several orientation mechanisms: main menu, internal search, sitemap, and breadcrumb. The goal is to cover different navigation profiles. A regular visitor uses the menu; a casual visitor searches by keyword; an assistive technology user relies on structural markers (headings, ARIA landmarks).
A sitemap compensates for the limitations of the navigation menu. A three-level menu mechanically hides pages beyond that. The sitemap makes them visible in one view, without additional interaction.
On a thematic content site, this flattened view has a second benefit: it reveals editorial density by category. If one category displays three pages and another displays forty, the imbalance is striking. It is a useful indicator for deciding where to produce additional content or merge underdeveloped sections.
Breadcrumbs and explicit page titles
The breadcrumb trail complements the sitemap by providing a permanent positional context. For it to be usable, each segment must repeat the exact title of the parent page, not a shortened or rephrased label.
We frequently observe breadcrumbs that are inconsistent with the structure displayed in the sitemap. This discrepancy disorients the visitor and harms readability for screen readers. Each level of the breadcrumb trail must correspond to a real node in the sitemap.
Testing the structure using a keyboard and screen reader
A visually clear structure can remain inaccessible if keyboard navigation is faulty. Recent W3C recommendations emphasize several technical criteria that we detail here.
- The visible focus must accompany each link in the sitemap: without a position indicator, a keyboard user does not know where they are in the list
- Repetitive navigation blocks (header, footer) should be bypassable by a skip link, allowing direct access to the sitemap content
- Section titles (Hn tags) structure reading for assistive technology: a sitemap without a hierarchy of titles presents as an undifferentiated continuous flow
- Navigation consistency requires that the order of links remains the same from one page to another, including the sitemap
Testing these points does not require expensive tools. The Tab key is sufficient to check the keyboard path. A free screen reader (NVDA on Windows, VoiceOver on macOS) allows for checking the rendering of titles and links.
Discrepancy between theoretical structure and actual visitor paths
Card sorting is a method that involves asking users to group content by affinity themselves. The results often reveal a gap between the organization defined by the editor and the expected classification logic from visitors.
On a site like Mes Rêves, whose content is organized around dream-related themes, categories may seem obvious to the site creator but counterintuitive to a new visitor. The sitemap then serves as a diagnostic tool: if a user has to go through more than three clicks from the sitemap to reach common content, the depth is excessive.
Orphan pages and internal linking
An orphan page is an indexed page that is not linked by an internal link from another page on the site. The HTML sitemap can theoretically address this issue by listing all URLs. In practice, a sitemap does not replace contextual internal linking: search engines value links embedded in the body of content much more than an isolated link in a list.
We recommend using the sitemap to inventory pages, then checking that each receives at least one contextual link from an article or a parent category. Without this work, the sitemap remains a fragile safety net.
Exploring the structure of a site like Mes Rêves therefore involves three levels of reading: the static view of the HTML sitemap, functional testing with a keyboard and assistive technologies, and the analysis of the actual internal linking. Each of these levels covers an angle that the other two ignore.
