Header Tags

What header tags (H1–H6) are and how much they actually matter for SEO — how Google and Bing use them, multiple H1s, heading order, keywords, and accessibility.

First published: Jun 25, 2026 · Last updated: Jul 18, 2026 · Advanced
demand #28 in On-Page#142 in Technical SEO#194 on the site

Header tags are the HTML heading elements H1–H6 that structure a page's content. They're a real signal — Google has called a heading 'a really strong signal' about what a section is about, and uses H1s when generating SERP title links — but they matter far less than most SEO tools imply. Google says heading order doesn't affect rankings and multiple H1s are fine; Bing prefers one H1. In my study of over a million domains, missing and multiple H1s were both extremely common and both low-impact. Use semantic headings to organize content for readers and screen readers; don't treat them as a keyword-ranking lever.

TL;DR — Header tagsHeader tags (heading tags) are the HTML elements <h1> through <h6> that mark the structural hierarchy of a page's content. They help users, search engines, and screen readers understand what each section is about. (H1An H1 tag is the HTML `<h1>` element that marks a page's primary heading — the big visible headline at the top of the content. It helps users, search engines, and screen readers understand what the page is about.–H6) are a real but modest signal. Google has called a heading “a really strong signal” about a section’s topic and uses <h1> elements as an input when generating SERP title linksThe title tag is the HTML title element in a page's head that specifies the document's title. It's the primary source for the SERP title link and a confirmed light ranking factor — but since August 2021 Google doesn't always show it verbatim. — but it also puts heading order and heading count under “things you shouldn’t focus on,” says multiple H1sWhen a single web page contains more than one <h1> element. The HTML standard permits non-nested sibling h1s, and Google treats multiple h1s as a low-priority ranking concern — but nested h1 patterns are non-conforming, and one clear h1 is still the accessibility default. are fine, and Illyes confirmed order doesn’t affect rankings. Bing diverges: it wants one H1 and says more “diminishes the value.” In my million-domain study, missing H1s (59.5%) and multiple H1s (51.3%) were both rampant and both low-impact. Use semantic headings — not CSS-styled text — to structure content for readers, screen readers, passage ranking, and snippets. Don’t treat them as a keyword lever.

Evidence for this claim HTML heading elements h1 through h6 represent six levels of section headings. Scope: Document structure and accessibility semantics. Confidence: high · Verified: MDN: Heading elements Evidence for this claim Google can use heading elements as sources when it generates title links. Scope: Title-link generation, not evidence that heading order or count directly determines rankings. Confidence: high · Verified: Google Search Central: Title links

What header tags are (and aren’t)

Header tags are the HTML heading elements <h1> through <h6> — the semantic markers of a page’s content hierarchy. H1 is the top level (usually the page’s primary heading), and H2–H6 step down through sections and subsections. They render visibly in the body of the page, which is what separates them from the <title> element: the title tagThe title tag is the HTML title element in a page's head that specifies the document's title. It's the primary source for the SERP title link and a confirmed light ranking factor — but since August 2021 Google doesn't always show it verbatim. lives in the <head>, drives the browser tab and the SERP title link, and isn’t shown on the page. The H1 is on the page. They frequently overlap in wording, but they’re different elements with different jobs.

One terminology note worth getting right: the formal name in the HTML spec is heading elements. “Header tags” is the informal industry term for the same thing — and it’s the term Bing uses in its own docs. Don’t confuse either with the <header> element, which is an HTML5 landmark for a page or section’s header area (logo, nav), not a heading.

How to pick a level. The decision rule is structural, not visual: a heading’s rank should match where it sits in the page’s section nesting, not how large you want it to look. An H2 introduces a major section directly under the H1; an H3 introduces a subsection under that H2; and so on. If you’re choosing a heading level to get a particular font size, you’re using the wrong tool — use CSS for that and pick the rank based on the content’s actual position in the outline.

There’s also a spec-level detail worth knowing here: browsers don’t infer a heading’s rank from how deeply it’s nested inside <section> or <article> elements. HTML once had a proposed “outline algorithm” that would let, say, an <h1> nested two <section>s deep count as an effective H3 — but no browser ever shipped it, and the HTML spec formally removed it from the living standard in 2022. Practically: always use the explicit rank you mean (<h1><h6>), and don’t assume wrapping a heading in nested <section> elements changes its effective level. It doesn’t.

How Google uses headings

Headings are a genuine signal — just not a strong, standalone ranking lever. A few distinct things are going on:

  • Topic understanding. John Mueller put it directly: “A heading is a really strong signal telling us this part of the page is about this topic.” Crucially, he said the level doesn’t matter much — “whether you put that into an H1 tagAn H1 tag is the HTML `<h1>` element that marks a page's primary heading — the big visible headline at the top of the content. It helps users, search engines, and screen readers understand what the page is about. or an H2 tag or H5 or whatever, that doesn’t matter so much” — it’s the heading-as-signpost that helps. And he was equally clear that “headings on the page are not the only ranking factor that we have.”
  • Title link generation. Since the August 2021 title generation update, Google uses <h1> elements as one of the primary sources for generating the title link it shows in search results. So a clear, prominent, distinctive H1 reduces the odds Google rewrites your title in the SERP — and a missing or vague one raises them.
  • Passage rankingPassage ranking is a Google AI system that scores individual sections ('passages') of a page so a single page can earn multiple relevance scores for different queries. Google still indexes whole pages — only the ranking changed.. Clean heading + paragraph units give Google parsable segments it can identify and rank as distinct passages within a long page. A muddled or deeply nested structure makes that harder.
  • Featured snippets. Heading-delimited content — H2/H3 questions, numbered steps — is exactly what Google extracts into list and table snippets. Structure feeds those.

Now the deflating part. Google’s SEO Starter Guide files heading guidance under “things we believe you shouldn’t focus on.” Two lines from there matter:

“Having your headings in semantic order is fantastic for screen readers, but from Google Search perspective, it doesn’t matter if you’re using them out of order.”

“There’s also no magical, ideal amount of headings a given page should have. However, if you think it’s too much, then it probably is.”

Gary Illyes reconfirmed the order point in the July 2024 SEO Office Hours, and explicitly sided with the Starter Guide over the SEO tools that flag heading order: “from Google Search perspective, it doesn’t matter if you’re using them out of order,” adding that “the web at large isn’t valid HTML code, so Google Search rarely relies on the semantic meanings hidden in the HTML specification.” When a non-Google tool flags something, he noted, “that doesn’t make it relevant for Google.”

Multiple H1s: Google is fine, Bing isn’t

This is the cleanest place where the two engines actually diverge, so it’s worth stating plainly.

Google has said this about as many ways as it can. Mueller: “You can use H1 tags as often as you want on a page. There’s no limit, neither upper or lower bound,” and “your site is going to rank perfectly fine with no H1 tags or with five H1 tags.” In the “Ask Google Webmasters” video: “Our systems don’t have a problem when it comes to multiple H1 headings on a page. That’s a fairly common pattern on the web.” He also said the systems will work with “styled pieces of text without semantic HTMLSemantic HTML is the practice of using elements like <main>, <article>, <section>, <nav>, <header>, and <aside> to describe what content is, not just how it looks. It helps search engines and assistive tech identify a page's main content more reliably, but it isn't a ranking factor on its own. at all.” So from Google’s side, H1 count is a non-issue for rankings.

Bing takes the opposite stance. Its long-standing webmaster guidance recommends one H1 per page and says using more “diminishes the value of the <h1> tag” — the logic being “there’s supposed to be only one big idea per page.” Bing Webmaster ToolsMicrosoft's free portal for monitoring and improving how a site appears in Bing search — the peer to Google Search Console, plus IndexNow instant indexing, richer backlink data, and keyword volumes. Because Bing's index also feeds Microsoft Copilot, it doubles as a window into AI-search visibility. actively flags multiple H1s, though only as a low-priority Notice, not an error. So if you care about Bing, leaning toward a single H1 is the safer default.

Where they agree: one H1 per page is the right call for accessibilityWeb accessibility means designing sites so people with disabilities can use them, per the W3C's WCAG guidelines. It overlaps with SEO in specific, checkable ways — alt text, heading structure, descriptive link text, captions, and page speed all serve both audiences — but Google has said accessibility itself is not a ranking factor, and most WCAG success criteria (keyboard focus order, ARIA live regions, form labels) have no SEO effect at all., regardless of what either engine does for rankings. That’s the real reason the “one H1” convention exists — screen reader users navigate by heading, and multiple H1s muddy the document outline.

What my data actually shows

When I studied over a million domains (1,002,165 of them) for the most common technical SEO issues, two heading findings stood out — and both are reassuring:

  • 59.5% of sites had pages with missing or empty H1 tags. My take at the time: missing H1s are a minor issue, especially if the text is wrapped in an H2 instead. Usually an easy fix, but I don’t expect much impact.
  • 51.3% of sites had pages with multiple H1 tagsWhen a single web page contains more than one <h1> element. The HTML standard permits non-nested sibling h1s, and Google treats multiple h1s as a low-priority ranking concern — but nested h1 patterns are non-conforming, and one clear h1 is still the accessibility default.. Multiple H1s are allowed in modern HTML and Google has said they’re not a problem. Worth confirming every H1-tagged heading is actually page-relevant, but I don’t treat it as significant.

In other words: the two heading “issues” your audit tool screams about loudest are present on more than half of the entire web, and neither is a meaningful ranking concern. I’ve made the point bluntly before — when someone claims an H1 change lifted their rankings, it was almost always because they simultaneously added keyword-rich copy high on the page (prominence), not the H1 element itself. You could swap your H1s for a styled <span> and the ranking impact would be roughly nothing.

That said: don’t read this as “headings are useless.” They’re useless as a keyword ranking trick. They’re genuinely useful as structure — for readers, for screen readers, and for the snippet/passage machinery above.

Keywords in headings

Use them — naturally. A heading should describe the section it sits above, and if your target topic is what the section is about, the keyword shows up on its own. The thing to avoid is forcing exact-match phrases into every heading. The correlation data points the other way: top-ranking pages are, if anything, slightly less likely to cram exact-match keywords into the H1. And remember Mueller’s framing — the heading’s job is to signal “this part of the page is about this topic,” not to be a keyword slot.

Also: the H1 and the title tag don’t need to match. Illyes, asked exactly this, said “No, just do whatever makes sense from a user’s perspective.” Some overlap in wording is natural; an exact match is not a requirement.

Semantic headings vs. CSS-styled text

This is the gap most heading advice skips. A heading that looks like a heading isn’t necessarily a heading. <div class="section-title">Big Bold Text</div> can be visually identical to an <h2>, but it carries no structural meaning — no signal to crawlersA crawler — also called a spider or bot — is an automated program that fetches web pages, extracts their links, and queues new URLs to visit. Search engines use crawlers to discover and download content for their index., and nothing for a screen reader to navigate by. Google can sometimes read visually prominent text (it’s one of the inputs for title link generation), but for general heading signals, passage rankingPassage ranking is a Google AI system that scores individual sections ('passages') of a page so a single page can earn multiple relevance scores for different queries. Google still indexes whole pages — only the ranking changed., and accessibility, you want real semantic <h> elements. If you’re on a CMSA content management system (CMS) is software that lets users create, manage, and publish digital content — like blog posts and pages — without writing raw code. WordPress, Drupal, and Joomla are the most common open-source CMS platforms., this is often controlled by the theme/template, not the editor — a heading you type in the editor might render as a styled paragraph depending on the template. Worth checking.

Heading order and accessibility

Two separate questions that constantly get merged:

  • For Google rankings: order doesn’t matter (Starter Guide + Illyes, above). Skipping from H2 to H4 won’t hurt your rankings. Martin Splitt’s framing is useful here — if you structure your content differently, “you didn’t structure it better, you didn’t structure it worse, you just structured it differently.”
  • For accessibility: order matters a great deal. Per WebAIM’s survey, the large majority of screen reader users navigate pages by heading, and a logical H1 → H2 → H3 outline is what makes that work. W3C’s own heading guidance recommends against skipping levels for exactly this reason — it’s strong authoring guidance to follow. But it’s not automatically a WCAG failure by itself: whether a skipped level actually fails a criterion like 1.3.1 (Info and Relationships) depends on evaluating the specific page, not a blanket “any skip = fail” rule. Treat it as a real problem to fix, not a guaranteed conformance violation to cite.

So my practical rule: keep a clean, logical hierarchy for the accessibility win, and don’t lose sleep over the heading-order warnings your SEO tool fires — those are optimizing for a thing Google has explicitly said it doesn’t use.

TIP Inspect the outline without turning every jump into an SEO error

This checker flags the exact H1-to-H3 jump and preserves the important caveat: Google does not require a perfect numerical sequence. Review it for document clarity and assistive navigation.

Run the page through my heading checker to see the rendered outline, empty headings, skipped levels, and fragment-link readiness together. Heading Structure Checker Free

  1. Review the page outline in DOM order and locate the exact level jump.
  2. Decide whether the deeper heading is genuinely nested or only styled to look smaller.
  3. Fix the semantic level when it improves reader and screen-reader navigation; do not forecast a ranking lift from renumbering alone.
A skipped level is a structural review prompt, not an automatic Google ranking failure.

A note on JavaScript-rendered headings

If your headings only appear after client-side JavaScript runs, they’ll generally still get indexed — but on Google’s two-phase render cycle, which can lag. And headings that only appear after a user interaction (clicking a tab, expanding an accordion) may not be indexed at all, because Google doesn’t click. For pages that need to rank, get your heading structure into the server-rendered or static HTML. (The renderingTurning HTML, CSS, and JavaScript into the final visual page and DOM. side of this lives in JavaScript SEOMaking sure search engines can crawl, render, and index content that depends on JavaScript..)

Best practices, condensed

  • Use one H1 per page as a default — mostly for accessibility and for Bing, since Google doesn’t care about the count.
  • Use H2/H3 to organize sections; H4–H6 only when the content genuinely nests that deep (most pages don’t).
  • Write descriptive headings that reflect the section; let keywords appear naturally rather than stuffing exact matches.
  • Use real semantic <h> tags, not CSS-styled divs.
  • Keep a logical hierarchy for screen readers — even though Google doesn’t require it.
  • Put critical headings in server-rendered HTML, not JS-only.
  • Don’t treat fixing headings on an existing site as a rankings play. Mueller again: “if you have an existing site, fixing this isn’t going to change your site’s rankings.” The wins are clarity and accessibility.

Where to go next

This hub is the overview of the whole H1–H6 family. The H1 tag gets its own deep dive as a sibling article in this cluster — H1-specific best practices, the one-H1 debate in detail, and how the H1 feeds title link generation. Multiple H1 tags gets its own deep dive too, focused specifically on the “does it hurt SEO?” question HTML5’s sectioning content raises. For the head element that’s often confused with the H1, see the title tag. The broader on-page picture lives in the On-Page SEO pillar.

Add an expert note

Pin an expert quote

New person? Create their unclaimed profile at /admin/experts/ → Pin a quote first.