A sitemap is not just an XML file submitted to search engines. It is an architectural document that outlines all accessible URLs, their hierarchy, and their relationships. When poorly designed, indexing bots miss deep pages, and users lose their bearings after the second click.
WCAG 2.2 Compliance and Sitemap: Requirements That Hierarchy Alone Does Not Cover
A sitemap can list all pages without being compliant. Since the publication of the WCAG 2.2 criteria by the W3C in October 2023, unmasked focus and drag-and-drop alternatives directly apply to navigation. A link present in the HTML sitemap but whose focus is hidden by a fixed banner remains non-compliant.
The European Accessibility Act, applicable since June 28, 2025, reinforces this constraint for e-commerce, online banking, and transport services. The structure must be keyboard usable and understandable with a screen reader.
We regularly observe sitemaps validated by automated audits but unusable for keyboard navigation. These tools detect missing tags or contrast errors, but not an illogical tab order or a keyboard trap in a modal window. Controls must include complete keyboard navigation paths, especially on filters, forms, and conversion tunnels.
Specifically, we recommend testing each critical path (search, cart, contact form) with Tab and Shift+Tab before considering the sitemap finalized. An accessible HTML sitemap serves both assistive technologies and search engines.
XML Sitemap, HTML Sitemap, and Hierarchy: Three Distinct Layers of a Sitemap
The confusion between these three elements produces shaky architectures. Each serves a specific role, and a well-structured sitemap articulates these three layers without redundancy.
The XML sitemap is aimed at bots. It declares the URLs, their update frequency, and their relative priority. It does not manage visual hierarchy or human navigation. Google treats it as an index, not as a directive.
The HTML sitemap is a page intended for users. It reproduces the hierarchy in the form of clickable links, organized by categories. On a site with several hundred pages, it is often the only place where a visitor can view all available content. By consulting the sitemap of La Top Liste, one can see how a category-based organization facilitates direct access to any section without going through the main menu.
The hierarchy, on the other hand, pertains to upstream design. It is the logical diagram that decides how many levels of depth exist, which pages are attached to which categories, and where secondary entry points are located. An HTML sitemap reflects the hierarchy, but it does not replace it.
- The XML sitemap must be synchronized with the CMS to reflect any page creation or deletion in real-time, or else it risks submitting 404 URLs to the engines.
- The HTML sitemap must adhere to a maximum depth of three levels to remain readable without excessive scrolling.
- The hierarchy must be validated by user testing before being finalized in the main navigation menu.
Evaluate a Sitemap by User Path and Not by Isolated Page
The most operational approach is to evaluate the compliance and effectiveness of a sitemap by path, not by page. A compliant page can belong to an unusable path.
Let’s take a concrete case: an e-commerce site where each product page meets accessibility standards. If the category filter system traps the user in a tabbing loop with no exit, even if the sitemap correctly lists the pages, the purchasing path is broken.
We recommend mapping critical paths before structuring the HTML sitemap:
- Search path: from the homepage to the results page, then to the target content page.
- Conversion path: from the category page to the contact form or payment tunnel.
- Return path: from a deep page to the parent category or homepage, without using the browser’s back button.
- Exploration path: from the HTML sitemap to any section in less than two clicks.
Each path must be tested with and without a mouse. If a path requires more than three clicks to reach any first-level page, the structure deserves to be reviewed.
Internal Linking and Sitemap: How Categories Fuel Indexing
A well-designed sitemap intentionally distributes crawl budget. Categories act as nodes in internal linking: they concentrate incoming links from the menu, footer, and HTML sitemap, then redistribute this authority to child pages.
Too many categories dilute the linking instead of strengthening it. We frequently observe sites with more than twenty first-level categories, resulting in an unreadable menu and an HTML sitemap that resembles a directory. It is better to group content into six to ten main categories, each containing subcategories if necessary.
The other common mistake is creating orphan pages, absent from the HTML sitemap and not linked by an internal link from another page. The XML sitemap may declare them, but without contextual links, engines assign them little value. Each page must be accessible from at least one text link in the body of another page, in addition to its presence in the sitemap.
Sitemap Update Frequency
An outdated XML sitemap sends bad signals. If deleted or redirected URLs are still listed, bots waste crawl budget on 301 or 404 responses. On a site where content evolves weekly, automatically generating the XML sitemap via the CMS is the only reliable option.
The HTML sitemap, on the other hand, can be updated less frequently. A quarterly review is sufficient for sites with a stable hierarchy. The goal is to ensure that each link points to an active page and that new categories are included.
Structuring a sitemap requires addressing technical indexing, human navigation, and accessibility compliance simultaneously. The three layers (XML, HTML, hierarchy) must be consistent with each other. A sitemap that does not reflect the actual architecture of the site loses its function for both engines and visitors.
