A sitemap serves two distinct functions that are often confused: guiding human visitors and feeding indexing bots. The first is an HTML page accessible from the footer, while the second is an XML file declared in the robots.txt file. When these two versions are poorly designed or absent, navigation degrades for users and indexing coverage decreases on the search engine side.
HTML Sitemap and XML Sitemap: Two Distinct Navigation Logics
The HTML sitemap lists the pages of a site in the form of clickable links, organized by categories. It is aimed at visitors who cannot find what they are looking for via the main menu or the internal search engine.
The XML sitemap, on the other hand, is a technical file intended for bots. It declares the URLs to be crawled, their update frequency, and their relative priority. Google and other engines use it to discover deep pages that the internal linking alone does not make visible.
Confusing the two means neglecting one or the other. An e-commerce site with hundreds of product pages needs a segmented XML sitemap by categories so that indexing covers the entire catalog. The same store also needs an HTML page where a lost visitor can find the complete structure, like the Double Portion sitemap that presents the hierarchy of its sections in one place.
Accessibility and Regulation: What the Sitemap Must Now Ensure
Since June 28, 2025, the European Accessibility Act applies to certain e-commerce, banking, transport, telecommunications, and media services. The HTML sitemap is no longer just a navigation convenience: it becomes a regulatory compliance element for the affected sites.
In France, the applicable reference remains RGAA 4.1.2, with an evolution towards RGAA 5 announced for the end of 2026. These references impose specific requirements on the navigation structure.
- Paths must be fully usable via keyboard, without relying on mouse hover to reveal submenus or links.
- The hierarchy of headings (H1, H2, H3) must remain consistent in the HTML sitemap so that screen readers can logically convey the hierarchy.
- The sitemap must not rely solely on visual elements (colors, icons without textual alternatives) to convey structural information.
The RGAA audit published by Saemes after an evaluation conducted in June 2026 illustrates a often underestimated point: actual compliance may remain low despite an apparently correct architecture. The existence of a sitemap is not enough if the links it contains are not accessible or if the page itself has markup defects.
Sanctions That Are Starting to Be Imposed
Arcom has begun to sanction breaches of digital accessibility in 2026. Sites that are structuring their navigation today should therefore avoid solutions that are difficult to evolve: menus relying solely on hover, inconsistent heading hierarchies, or paths that are impossible to navigate without a mouse.
Tree Structure: What Makes a Sitemap Truly Useful
A poorly structured HTML sitemap can contain all the pages of a site without actually helping anyone. The question is not completeness, but the readability of the proposed hierarchy.
Structuring pages (home, main sections, sub-sections) must appear in a clear hierarchical order. Secondary pages (legal notices, privacy policy) do not need to appear at the top.
The Case of Dynamic Content
Blogs, news sections, and product catalogs pose a specific problem. Should every article or product sheet be listed in the HTML sitemap? For a blog with a few dozen articles, the list remains manageable. For a catalog with several thousand references, it becomes unreadable.
The most common solution is to list categories and subcategories in the HTML sitemap, then delegate the details to the XML sitemap. The latter can be segmented (one file per product category, another for blog articles) and submitted via the Search Console.
XML Sitemap: Common Errors That Limit Indexing
Automatically generating an XML sitemap via a CMS like WordPress does not guarantee that it is correct. Several recurring errors reduce its effectiveness.
- Including URLs that return a 404 code or a 301 redirect. Search engines consider these declarations as noise and may slow down the crawling of valid pages.
- Declaring URLs blocked by the robots.txt file. The sitemap says “index this page,” the robots.txt says “don’t go there”: the contradictory signal harms coverage.
- Not updating the last modified date. If all URLs display the same date, search engines lose the freshness indicator of the content.
- Exceeding the limit of 50,000 URLs per file without creating a sitemap index. Overly large files may be partially ignored.
A well-maintained XML sitemap does not directly improve ranking. Its role is to facilitate the discovery and crawling of pages by bots, which is a prerequisite for any visibility in search results.
Automatic or Manual Generation
Automatic generation suits the majority of sites. WordPress extensions or CMS modules produce up-to-date sitemaps without intervention. However, regular manual checks are still necessary to spot obsolete URLs, orphan pages, or inconsistencies between the sitemap and the actual site structure.
For sites whose structure changes frequently (adding categories, removing seasonal pages), a quarterly check via the Search Console allows for detecting errors before they affect indexing coverage.
The sitemap, whether HTML or XML, remains a tool for maintenance as much as for navigation. A site whose sitemap accurately reflects the actual structure offers a consistent experience for visitors and a clean signal for search engines. Those who treat it as a formality miss out on a concrete lever, especially since accessibility obligations now make its proper design verifiable and sanctionable.
