Free SEO Tool · No Signup Required

Free Header Tag Checker

Paste one URL and see whether the page has an H1 with real words in it, whether it has only one, and whether that H1 says something the title tag does not. On articles it also looks for H2 sections and for question headings with no FAQ markup. The reason for every miss is right here.

Read-only. One fetch of the page from our server, the served HTML only, and nothing is written to your site.

  • No signup, no email wall
  • Same result every run
  • H1 count and text
  • H1 versus title tag
  • H2s on article pages
  • Free, no daily cap
What we checked

5 heading checks: 3 on every page, 2 more on articles. One GET of the URL as VerandBot/1.0, redirects followed, 12 second timeout, the served HTML only: no JavaScript is run. H1s are counted across the whole document. H2s and question headings are read from the main content, with the navigation, header, footer and sidebars stripped out.

How a row is decided

Each row is a fixed rule with one answer. The H2 row is one of three conditions inside the product's answer-structure check, read from that check's own detail text; its other two conditions are about the prose and stay off this card. A page counts as an article when it has Article schema, sits in a folder like /blog/, or shows two article signals such as a byline and a date. A homepage never does.

What it cannot see

It does not draw the H1 to H6 outline, flag a skipped level such as H2 to H4, or read H3 to H6, apart from counting question headings in H2 to H4. Empty headings are caught for the H1 only. It cannot tell a visible H1 from one hidden in a menu: both count. Headings a script adds after load are invisible to it. FAQPage schema is looked for in the first 100,000 characters of the HTML.

About this tool

How this heading checker reads a page.

One fetch, two readings of the same HTML, five questions. The card is the tool in motion on an example page, looped, and each step lights up while the card is doing it.

01

One request, the served page

The tool fetches the URL you pasted as VerandBot, follows redirects and gives up after twelve seconds. It reads the HTML the server sends. A heading that a script writes in after the page loads is not there to read.

02

Two readings of the same HTML

Every <h1> in the document is counted, including one tucked inside a hidden menu, and the first one's text is kept. Separately, the main content is isolated, with navigation, header, footer and sidebars stripped, and its H2s and question headings are read there.

03

Five questions, one answer each

Is there an H1 with text? Only one? Is the first H1 worded differently from the title tag? On articles: are there at least two H2 sections, and do three or more question headings come without FAQPage schema? Each is a fixed rule, so each answer is pass or fail.

04

The reason, in words

Every miss arrives with the check's own detail, such as how many H1s it counted. There is no heading score to chase: the card counts what passed and what did not, and says which rules did not apply.

Header tags, explained

Heading hierarchy: the outline a page hands to everyone who reads it.

Readers skim it, screen readers navigate by it, search engines and AI assistants use it to tell one section from the next. Here is what heading tags are, what Google actually says about them, and why regulated firms have more riding on them than most.

What H1 to H6 are

HTML has six heading elements, <h1> through <h6>. In SEO conversation they are called header tags or heading tags, and the two names mean the same thing. The number is a rank, not a size: an H1 names the page's subject, each H2 opens a major section under it, each H3 a subsection inside an H2, and so on down. Most pages never need anything below H3; a long guide might reach H4.

Read in order, a page's headings should work as its table of contents. If you copied only the headings into a document, a stranger should be able to say what the page covers and how it is organised.

<h1>Roth conversions, explained</h1>
  <h2>Who can convert?</h2>
  <h2>How the five-year rule works</h2>
    <h3>Conversions versus contributions</h3>
  <h2>What a conversion costs in tax</h2>
  <h2>Important disclosures</h2>

Header tags, the header element and HTTP headers

The word "header" covers three unrelated things, and search results mix them up constantly. Someone looking for a header checker is often after the third one.

TermWhat it isWhat it does
<h1> to <h6>Heading elementsTitle the page and its sections. These are what SEO means by header tags, and what this tool checks.
<header>A page regionWraps introductory content, usually the logo and navigation at the top of the page or the top of an article. It is not a heading and carries no rank.
HTTP headersResponse metadataLines the server sends before the page itself: status, content type, caching, security policies, X-Robots-Tag. Invisible on the page, and a different tool's job.

The overlap causes real mistakes. A theme that puts an H1 inside the site <header> region, usually wrapping the logo, gives every page on the site the same H1: the company name. The page's actual subject then arrives as a second H1, or as an H2.

The rules of a clean outline

Four habits make an outline work. One H1 that names the topic. No skipped levels: an H4 directly under an H2 tells a screen reader user that a level exists which does not. No empty headings: page builders leave behind headings with no text, which read aloud as "heading level two" followed by silence. Headings for structure, never for styling: if a line should look large but does not open a section, it needs a CSS class, not a heading tag.

The checker on this page tests the top of that outline: the H1, and on article pages, whether there are H2 sections at all. It does not yet walk H3 to H6 or flag a skipped level, so the rest of this list is for checking by eye. View the page source, search for <h, and read the matches in order.

One H1, and what Google actually says

Google's John Mueller has said more than once that several H1s on a page are not a problem for Google's systems, and Google's own SEO Starter Guide puts it plainly: headings in semantic order are "fantastic for screen readers, but from Google Search perspective, it doesn't matter if you're using them out of order." Anyone promising that header tags SEO work alone will move rankings is selling something Google has said it does not reward.

Verand flags more than one H1 anyway, for two reasons. One H1 is the clearest signal to a reader and to assistive technology of what the page is about. And a page with several H1s almost never chose them: they come from a logo, a menu label, a modal, a carousel or a template repeated on the page. The failing sample in the card above is our own homepage, where the animated product illustrations each carry an H1 and the count reaches ten. The fix there, as everywhere, is to demote the accidental ones.

The same thinking sits behind the H1-versus-title check. An H1 identical to the title tag is not an error. The title is what appears in a search result and the H1 is what greets the reader, so wording them differently lets each do its own job. Treat that row as a suggestion, not a defect.

Why bold text is not a heading

A line set in bold, or in a large font, looks like a heading to a sighted reader and is ordinary text to everything else. Screen readers do not list it in their headings menu. A browser's reader view does not treat it as a section break. Software that splits a page into sections has no reason to start a new one there. Visual weight and document structure are separate layers, and only the heading tag writes to the second one.

This is where regulated firms get caught. Disclosures are very often set as small bold text at the foot of a section: "Past performance does not guarantee future results," or "This is not legal advice." Because the line is not a heading and sits at the end of a block, it belongs structurally to whatever came before it, or to nothing. When a section is lifted out of the page on its own, the qualifying line is the easiest part to lose.

Headings as passage boundaries for AI answers

AI assistants rarely quote a whole page. Search and answer systems commonly break a page into passages and work with the passage that best matches the question, and headings are the natural place for those breaks: the heading tells the system, and the eventual reader, what the passage under it is about. A page with no headings is one long passage. A page whose headings are vague ("Overview," "More info") hands every passage an unhelpful label.

An example. A financial advisor's article has an H2, "Can I convert a traditional IRA to a Roth?", followed by two paragraphs that answer it, and a closing bold line noting that conversions are taxable and the reader should consult a tax professional. If that answer is extracted as a passage, the claim travels and the caveat may not, because nothing in the structure ties it to the answer. Two fixes help: put the essential qualifier inside the answer itself, and give standalone disclosures their own heading so they form a section of their own. Neither guarantees how any assistant uses the page. Both make the page say what it means in its structure, not only in its styling.

Headings, screen readers and accessibility

Screen reader users navigate by heading more than by any other means. One keystroke jumps to the next heading and another opens a list of every heading on the page, which works as a table of contents for someone who cannot skim. The Web Content Accessibility Guidelines cover this in success criterion 1.3.1, Info and Relationships (Level A), which asks that structure shown visually is also present in the markup, and in 2.4.6, Headings and Labels (Level AA), which asks that headings describe their topic. Other checkers explain this well; it is table stakes.

What is different for a law firm, a medical practice or a financial advisor is the exposure. Web accessibility demand letters and lawsuits reach professional practices in the US, and missing or broken heading structure is among the first things an automated accessibility scan reports, because it is so easy to detect. This is risk awareness, not legal advice: a clean heading outline does not make a site accessible on its own, and a missing one is not by itself a violation. It is simply one of the cheapest defects to find and fix before someone else finds it.

Common mistakes

  • The logo as the H1. Every page gets the company name as its main heading and the real topic drops a level.
  • A second H1 you cannot see. Mobile menus, modals, cookie banners and slider templates often carry their own H1. Hidden or not, it is in the HTML, and this checker counts it.
  • An H1 that is an image. A logo or banner image inside an H1 gives it no text. This checker reads the H1's text and reports an H1 with no words as missing.
  • Skipping from H2 to H4 for size. The heading level was picked for how it looks. Style the H3 instead.
  • Headings used as call-outs. "Call us today" or a phone number set as an H2 opens a fake section and pollutes the outline.
  • Empty headings from page builders. A deleted block leaves its heading tag behind, invisible on the page and announced by a screen reader.
  • Question headings with no FAQ markup. Three or more question-style headings on an article with no FAQPage schema is flagged on the article card here. The schema labels the questions and answers the page already shows; it adds nothing the page does not say. Google stopped showing FAQ rich results in Search in May 2026 and has since removed its FAQPage documentation (Google's changelog), so treat this as clarity for machines, not a search feature.
  • Headings written in by JavaScript. If the headings exist only after scripts run, any tool that reads the served HTML, this one included, sees none. So do some crawlers.

What a good outline looks like

One H1 that says what the page is about, in words a reader would use. H2s that each name a question or a claim the section answers, so the list of H2s alone reads as a summary. H3s only where an H2 genuinely splits. Disclosures under their own heading, or written into the answer they qualify. No headings in the navigation, the footer or the sidebar unless they title a real region. Then run a heading structure checker after every redesign or theme update, because templates are where extra H1s come from.

Why this one

Why choose Verand's Header Tag Checker?

Six things that are true of this tool, each one backed by a line in the code that runs it.

No signup, no email wall

The request carries a URL and nothing else. There is no account, no session and no database behind the tool, so there is nothing for us to keep about you.

The reason for every miss, on the page

Each failed rule comes back with the check's own detail, such as how many H1s it counted. Nothing is held back for a report, a call or an upgrade.

Deterministic

Tags are counted and compared by fixed rules. No model reads the page, so the same HTML gives the same answer every time.

The product's own checks

These rows come from the same audit Verand's deep crawl runs on every page of every customer site every two weeks, called on one URL. Not a lighter demo version.

Names what it cannot see

No full outline, no skipped-level test, no H3 to H6 beyond question headings, nothing a script adds. The card says so beside the result rather than in a footnote.

$0, no daily cap

A run is one fetch of one page, so it costs nothing and is never metered. The one limit is a courtesy to the sites being fetched: 20 checks a minute per visitor.

Questions

Frequently Asked Questions About the Header Tag Checker

Heading order, skipped levels, Google, bold text, screen readers, and the three things called a header.

What is the correct order for H1 to H6 headings?

One H1 that names the page's subject, then H2s for its main sections, H3s for subsections inside an H2, and so on, each level nested inside the one above it. Going back up is fine: after an H3 the next section can open with an H2. Read top to bottom, the headings should work as the page's table of contents. This checker tests the top of that order, the H1 and on articles the presence of H2s; it does not yet walk the levels below.

Is it bad to skip a heading level, like H2 to H4?

For Google Search, no: its SEO Starter Guide says heading order does not matter from a Search perspective. For people using screen readers it is a small but real problem, because a jump from H2 to H4 implies a level that does not exist, and accessibility checkers commonly flag it. Skipped levels usually mean a heading was chosen for its size, so the fix is to use the right level and restyle it. This tool does not flag skipped levels today, and it is not an H2 tag checker in that sense: it confirms that an article has at least two H2 sections.

Does heading order affect Google rankings?

Google says it does not. Its SEO Starter Guide describes semantic heading order as fantastic for screen readers and says that from a Google Search perspective it does not matter if headings are used out of order, and Google's John Mueller has said several H1s on a page are not a problem for its systems. Fix your headings for readers, screen reader users and clarity; do not expect a ranking change from tidying the nesting alone.

Why doesn't bold text count as a heading?

Because bold is a visual style and a heading is a structural element. A line in bold looks like a heading to a sighted reader but is ordinary text to a screen reader, a browser's reader view, and any software that splits a page into sections. Only an h1 to h6 tag puts a line into the outline. That matters most for disclosures, which are often set as small bold text: outside the outline, they are easily separated from the claim they qualify when a section is read on its own.

How do headings affect screen readers and accessibility?

Headings are how screen reader users move through a page: one key jumps to the next heading and another lists them all. WCAG success criterion 1.3.1 asks that visual structure also exists in the markup, and 2.4.6 asks that headings describe their topic. For regulated professional firms, broken heading structure is also one of the easiest defects for an automated accessibility scan to find. That is risk awareness, not legal advice: headings alone do not make a site accessible.

What is the difference between header tags and the HTML header element?

Header tags, in SEO language, are the heading elements h1 to h6, which title a page and its sections. The header element is a region of the page, usually holding the logo and navigation, and carries no heading rank of its own. HTTP headers are a third thing entirely: metadata the server sends before the page, such as status codes and caching rules. This tool checks the first. A theme that wraps its logo in an H1 inside the header element is a common source of a second H1.

After the check

Fix the outline. Then give it something worth citing.

A clean heading structure is the frame, not the finish. Verand writes articles from your own expertise and credentials, and runs its heading checks on every draft before you see it, so Google and the AI assistants have something of yours to name. It then tracks where you rank and where ChatGPT, Gemini, Google AI Overviews, Google AI Mode, Perplexity and Claude mention you, and gates every draft so a claim your regulator would not allow never publishes.

Verand

Content built to rank in Google and get cited by ChatGPTPerplexityGeminiClaude, with every claim checked before it goes live.

support@verand.ai

© 2026 Verand. All rights reserved. TermsPrivacyAI policyAccessibilitySecurity
Not legal advice. Compliance packs are researched from the regulators' own text and tested by Verand, not reviewed by a licensed attorney.