A poorly designed sitemap helps no one. Neither search engines, nor visitors, nor assistive technologies. We regularly observe sitemaps that list hundreds of URLs without logic, mixing orphan pages, redirects, and duplicate content. The result: a file that Googlebot crawls without deriving any useful signal, and an HTML page that no one consults.
XML Sitemap: Filter URLs Before Submission
An XML sitemap should only contain canonical, indexable URLs that respond with HTTP 200. This is the foundation, yet the majority of files automatically generated by CMSs include redirected pages, pagination variants, or URLs marked with noindex.
This noise degrades the quality of the signal transmitted to crawlers. When a sitemap contains several thousand entries, a significant fraction of which returns 301 or 404 errors, the crawl budget is wasted on unnecessary resources.
We recommend a quarterly audit of the XML sitemap with a dedicated crawler. The audit focuses on three points:
- Check that each URL responds with 200 and is not redirected to another address
- Ensure that the canonical tag of each page points to itself (not to a third-party URL)
- Exclude any page with a
noindexdirective in its HTTP header or meta tag
This cleanup transforms the sitemap into a reliable index. Search engines find exactly what the site wants to index, without ambiguity. A concrete example of a well-organized structure: the sitemap of Le Bon Sens groups its content by thematic categories, which facilitates both human reading and automated crawling.
HTML Sitemap and Accessibility: A Second Navigation Path
The HTML sitemap is often relegated to the footer, treated as a relic from the 2000s. This is a mistake. Accessibility recommendations advocate providing at least two general means to locate a page on a site: main menu, internal search, breadcrumb, or sitemap.
The European Accessibility Act reinforces this requirement for online commerce. The obligation covers menus, filters, pop-ups, and checkout steps. A site whose navigation relies solely on a complex mega-menu excludes users who navigate via keyboard or screen readers.
Designing the HTML Sitemap for Keyboard Navigation
An accessible HTML sitemap is not just a simple list of links. Each link must have an explicit title (no “click here”), and the page must be fully navigable without a mouse. This involves a logical tab order, a visible focus on each interactive element, and the absence of keyboard traps.
Compliance is not verified solely with an automated tool. Key journeys (accessing a product sheet, reaching the contact page, returning to the homepage) must be tested manually with a keyboard. A visual audit does not detect a link whose focus is hidden by CSS.
Heading Hierarchy in the Site Structure
The structure of Hn headings on a page directly contributes to its navigability for screen readers. A level jump (h1 followed by an h4) breaks the flow and makes the content difficult to understand, even when the visual menu seems clear.
For the HTML sitemap, we apply a simple rule: one h2 per main category, one h3 per subcategory. No h4 unless there is a real three-level hierarchy. This hierarchy reflects the structure of the navigation menu and allows assistive technologies to jump directly to the relevant section.
Consistency Between Menu, Sitemap, and Page Structure
A common pitfall: the main menu displays titles different from those in the sitemap, which themselves do not match the title tags of the pages. This inconsistency disorients the user and dilutes the internal linking.
Each entry in the sitemap must exactly match the title used in the menu. The link text, the title of the target page, and its h1 must share the same lexical field. This consistency strengthens both the user experience and the semantic signal sent to search engines.
Website Structure: Depth and Internal Linking
No useful page should require more than three clicks from the homepage. This is an information architecture rule, not just an SEO tip. A well-designed sitemap makes this depth visible and verifiable.
When a page is five levels deep, it receives little internal link juice and remains under-visited. The HTML sitemap allows for the identification of these buried pages and for deciding whether they deserve a direct shortcut or a restructuring of the hierarchy.
- List all pages more than three levels deep and assess their usefulness
- Create cross-links between related categories rather than multiplying sub-levels
- Update the sitemap with each addition or removal of a page, not just during redesigns
A static sitemap, unchanged since launch, loses its value within a few months. Content evolves, pages are removed, others appear. The XML sitemap and the HTML sitemap must reflect the actual state of the site, not a dated photograph of its initial structure.
Regular maintenance of the sitemap, whether XML or HTML, remains the best indicator of the technical rigor of a web project. A site with a clean and up-to-date sitemap signals to both search engines and visitors that navigation has been thoughtfully designed, not merely endured.
