// HOW-TO · TECHNICAL SEO

How to build an HTML sitemap for content discovery (2026)

Build an HTML sitemap that helps users and crawlers reach your content: what to list, how to structure topic hubs, and why it never replaces XML.

Last verified · 2026-10-10 · by Moe Ameen

An HTML sitemap is a plain page on your site that links to your important content, grouped so a visitor can see the shape of the site and drill down to what they want. On Google's Search Off the Record podcast, John Mueller described it to Martin Splitt as "a map of your website for users" — a navigation aid, not a search-engine file. That distinction is the whole point: an HTML sitemap is built for people first, and the discovery benefit for crawlers is a side effect of being a page full of real internal links.

It is not a replacement for your XML sitemap. The XML file is a strict, machine-readable list you submit to Google Search Console; the HTML page is a human-facing index you link from your footer or navigation. Done well, an HTML sitemap gives visitors a fast route to your category and hub pages and gives crawlers another path to follow links — done badly, it becomes an unwieldy wall of every URL on the site that helps no one. This walks the version worth building.

The steps

Decide whether you actually need one. An HTML sitemap earns its place on a site with real depth — many categories, hubs, or sections a visitor might struggle to navigate. If your main menu and on-site search already get people where they need to go, you may not need one at all. Mueller has made the sharper point before: if people are relying on the sitemap page to find things, that is a signal your normal navigation is failing. Build it to complement good navigation, not to paper over bad navigation.
Map your top-level categories and hub pages first. Start from structure, not URLs. List your main sections — the category or topic-hub pages a visitor would recognize — and treat those as the backbone. This is the layer Mueller specifically recommended including so users can drill down from a broad area to a specific page. If you cannot name your top-level hubs in a sentence each, fix your site's information architecture before you build the sitemap page.
Link to what matters — not every page. Resist the urge to dump every URL. Mueller's example was blunt: an ecommerce site should not list every product on its HTML sitemap. Include the pages that help someone navigate — hubs, key categories, cornerstone articles, important landing pages — and leave out the long tail that nobody browses to from a sitemap. A focused page of 50 useful links beats an exhaustive page of 5,000 that no visitor reads.
Group links so a visitor can scan and drill down. Organize the links under clear headings that mirror your top-level categories, so the page reads as a map rather than an undifferentiated list. Use descriptive, human anchor text — the words a visitor would search for — because that text also gives crawlers context about the linked page. A scannable, grouped layout is what turns a link list into an actual navigation aid.
Make it reachable from users and crawlers. Link the HTML sitemap from a persistent location — most commonly the site footer, which is where SEOs historically placed it so every page passed link equity and a crawl path to it. A sitemap page nobody can reach helps nobody. Keep the page itself a normal, crawlable HTML page: real anchor links, no login wall, not buried behind JavaScript that a crawler cannot render.
Keep your XML sitemap as the primary discovery file. The HTML sitemap supplements, it does not replace. Maintain a current XML sitemap and submit it in Google Search Console — that is the strict, machine-readable file search engines process directly, and it carries signals (like last-modified dates) an HTML page does not. Think of XML as the file you hand the search engine and HTML as the page you build for the visitor; a healthy site has both doing their own job.
Prune and maintain it as the site grows. An HTML sitemap rots the same way any hand-curated page does — links break, sections get renamed, new hubs appear. Review it on the cadence you ship new sections, remove dead and redirected links, and add new hubs as they launch. A stale sitemap page sends users and crawlers to 404s, which is worse than not having one. If the page starts ballooning past what a human would scan, that is the signal to tighten it back to hubs.

Common gotchas

Treating an HTML sitemap as an XML replacement. They are different tools — XML is the machine file you submit to Search Console; HTML is the user-facing page. You need both.
Listing every page. An exhaustive dump (every product, every tag) makes the page useless to visitors and dilutes its value — include hubs and key pages, not the long tail.
Relying on it to fix discovery. If users and crawlers only reach certain pages via the sitemap page, the real problem is weak internal navigation and linking — fix that first.
Thin or duplicated anchor text. "Click here" or the bare URL gives neither users nor crawlers context; use descriptive anchors that name the destination.
Letting it go stale. Broken and redirected links accumulate fast; a sitemap page that points to 404s actively hurts the experience it was built to improve.
Burying it or rendering it with JavaScript a crawler cannot execute. If the links are not in crawlable HTML and the page is not linked from anywhere, it does nothing for discovery.
Expecting a ranking boost from the page itself. Its value is navigation and an extra crawl path — it is not a ranking lever, and building it for that reason misses the point.

Where Kompozy fits

An HTML sitemap is a navigation layer over a library of content — and the hard part is rarely the sitemap page itself, it is producing enough on-brand, organized content to make the map worth drawing. That is where [Kompozy](/) comes in. It is a full AI content generation and multi-platform publishing engine, so the hubs and cornerstone pages your sitemap groups together are things it actually produces, not just schedules.

The concrete workflow: Kompozy generates Blog Articles from a single source — a talk, a product line, a cluster of questions — and publishes them to your blog via its blog destinations (GHL Blog, WordPress, or a Custom Webhook). Run that on a cadence and you accumulate exactly the structured library an HTML sitemap exists to organize: topic hubs, supporting articles, landing pages. Because a [Persona Brief](/glossary/persona-brief) governs voice across every piece, the pages group cleanly under coherent themes instead of reading as a random pile — which is precisely what makes Mueller's "drill down from a top-level category" structure possible to build.

It goes wider than the blog, too. Kompozy generates across [18 formats](/glossary/output-buckets) and fans them to eight social platforms plus blog and email behind [Autopilot](/glossary/autopilot) and a per-post review gate, so the same source feeds your site's content depth and its off-site footprint at once. The sitemap page and the XML file you submit to Search Console are yours to maintain — this is not an SEO-plumbing tool — but Kompozy produces the volume of organized, on-brand content that makes an HTML sitemap something visitors actually use. A solo creator or small site fits Starter ($199/mo, 5,500 credits); a publisher running a continuous content library fits Pro ($499/mo, 18,000 credits); Enterprise is custom for multi-brand operations.

Frequently asked questions

What is an HTML sitemap?

It is a regular page on your website that links to your important pages, grouped so visitors can see the site's structure and find what they want. Google's John Mueller called it "a map of your website for users." Because it is a page full of real internal links, crawlers can also follow those links, so it gives discovery a modest, secondary assist — but its primary job is human navigation.

What's the difference between an HTML and XML sitemap?

Purpose and audience. An XML sitemap is a strict, machine-readable file you submit to search engines in Google Search Console; it lists URLs with metadata like last-modified dates. An HTML sitemap is a human-facing web page of grouped links you build for visitors. XML is for search engines, HTML is for people, and they are complementary — an HTML sitemap does not replace the XML file.

Do HTML sitemaps help SEO?

Indirectly and modestly. They are not a ranking factor and not a substitute for your XML sitemap. Their SEO value is that they are a page of internal links, so they can give crawlers another path to discover pages and pass context through anchor text. The real win is user navigation; any crawl benefit is a side effect of building a genuinely useful page.

Should I list every page on my HTML sitemap?

No. Mueller's guidance is to include the pages that help navigation — top-level categories and hubs — and leave out the rest; he specifically said an ecommerce site should not list every product. A focused page organized around useful paths serves visitors far better than an exhaustive wall of links, which is unwieldy and ignored.

Does Google recommend HTML sitemaps?

Google frames them as optional and user-focused, not required. Mueller described the HTML sitemap as a map for users and was clear it does not replace an XML sitemap. If your site is large and hard to navigate, a well-built HTML sitemap helps; if your navigation and search already work well, you may not need one — and over-relying on it can signal that your navigation is weak.