Free tool · On-page SEO
Free h1 checker
To check a page's h1, read how many there are, how long the first one is, and how much it overlaps the title tag. Enter a page below to see every h1 with its length, an empty or image-only h1 called out as such, the full heading outline with any skipped levels, and the threshold behind each finding.
Free. The h1 audit needs no account; the answer-engine heading layer needs a free sign-in.
AEO depth layer
See the answer-engine heading layer
The answer-engine heading layer is computed from your own page and costs nothing to run. We ask for a free account because the expensive tools on this shelf stay free, and because it lets us tell you when your result changes. No card, no trial clock.
- Whether your h1 is shaped like a question an answer engine can match to a real query, and which of your h2s already are.
- Whether the first two sentences after the h1 actually answer it, which is what decides if there is anything self-contained to quote.
- The exact sentence an engine would lift from the page as the answer, so you can judge it rather than guess.
Signing in saves the details you entered above so we can follow up about your results. See our privacy policy for what we keep and why. The H1 Checker result you already have stays free either way. Privacy policy.
Every check, and the threshold behind it
These are the rules this tool applies, with the number that triggers each one. They are published here because a score you cannot interrogate is not worth having, and because two of them are our judgement rather than anyone's published limit.
| Check | Threshold | Level | Why |
|---|---|---|---|
| h1 present | zero h1 elements | Problem | Nothing states the page's subject in the strongest place available. |
| h1 has text | an h1 with no text, including an image-only h1 | Problem | An engine reads the text, not the picture. Most checkers report this as no h1 at all, which sends you looking for the wrong fix. |
| Single h1 | more than one h1 | Worth fixing | Valid HTML5 and tolerated by search engines. The cost is ambiguity, not a penalty. |
| h1 length | under 15 or over 70 characters | Worth fixing | Our judgement, not a published limit. Google does not state an h1 length. |
| Title agreement | under 30% word overlap | Worth fixing | Measured on meaningful words with common words removed. A mismatch reads as uncertainty about the subject. |
| Keyword in h1 | target keyword absent | Worth fixing | Word-boundary match, so "monitor" does not count as "monitoring". Only checked when you supply a keyword. |
| Outline order | a skipped heading level, or a heading before the h1 | Worth fixing | A gap in the outline leaves the structure ambiguous for anything reading it as a document. |
How this compares to a typical free h1 checker
Where the usual free checkers are equal, the table says so.
| Check | This tool | Typical free checker |
|---|---|---|
| h1 count and text | Yes | Yes |
| h1 character length | Yes | Usually |
| Empty or image-only h1 named as such | Yes | Reported as no h1 |
| Title overlap as a percentage | Yes | Rarely, and usually as a pass or fail |
| Keyword matched on word boundaries | Yes | Substring match, so partial words count |
| Full heading outline with skipped levels | Yes | Sometimes an outline, rarely the violations |
| The threshold behind each finding shown | Yes | No |
| Multiple h1s treated as a warning, not an error | Yes | Often a critical error |
| Answer-engine heading read | Yes, with a free account | No |
What is in the answer-engine heading layer
A heading does two jobs now. It tells a reader what the page is about, and it gives an answer engine something to match a question against. The second job has different rules, and the layer behind the free sign-in reports on it.
- Whether the h1 is shaped like a question, and how many of the h2s are. Question-shaped headings are the units an engine quotes, because they map onto what someone typed.
- Whether the first two sentences after the h1 restate its subject. A heading that asks something and an opening paragraph that talks about the company leaves nothing self-contained to lift.
- The exact sentence an engine would most likely quote, so you can read it and decide whether you would want that sentence representing the page.
It is a structural read of your own page, not a model call, and it measures whether an engine could extract an answer rather than whether one did. For what the engines actually say, the next step is the AI visibility checker. To confirm they are allowed to read you at all, the AI crawlability checker reads your robots.txt bot by bot, and the llms.txt generator builds the file that tells them what to cite.
How to check the h1 on a page
Step 1
Enter the page address
Paste the page you want to audit, for example acme.com/pricing. A public domain, not an IP address or a staging host that only you can reach.
Step 2
Read the overlap number before the verdict
The overlap percentage between the h1 and the title is the finding worth your attention. A low number means the page is ambiguous about its own subject, which no amount of h1 tidying fixes.
Step 3
Fix the failures, then judge the warnings
Only a missing or empty h1 is a failure. Multiple h1s and length warnings are judgement calls, and the tool shows the threshold so you can decide.
Step 4
Check whether an engine could quote the page
The heading layer reports whether your h1 is question-shaped and whether the opening actually answers it. Then use the AI visibility checker to see whether engines recommend you at all.
When you outgrow this tool
A heading audit tells you whether one page could be quoted. Answer Radar tells you which pages the engines actually cite for your buyers' questions, and which competitor they name instead.
The h1 audit is free and complete on its own. Linkeddit is what you use when checking one page by hand stops being enough and you need the whole site watched every week without you.
FAQ
H1 Checker questions
What an h1 actually does, what the thresholds here mean, and what a server-side read cannot tell you.
How many h1 tags should a page have?
One is the safe default, because the h1 is the clearest single statement of what a page is about. More than one is valid HTML5 and search engines cope with it, so this tool reports multiple h1s as a warning rather than a failure. The real cost of several h1s is not a penalty, it is ambiguity: each one claims to be the subject of the page, and only one can be.
Does having multiple h1 tags hurt SEO?
Not directly, and any tool that tells you it is a critical error is overstating it. HTML5 permits an h1 per sectioning element and search engines have handled that for years. What does cost you is a page whose headings disagree about its topic, or a layout that uses h1 for styling so the actual subject is buried in an h2.
What is a good h1 length?
This tool warns below 15 characters and above 70. Those are not rules from Google, which publishes no h1 length limit; they are the range where a heading is long enough to contain the phrase someone searched for and short enough to read as a heading rather than a sentence. The exact count is shown either way so you can overrule it.
Should the h1 and the title tag be identical?
No, and identical is often worse. The title competes in a result listing and can carry a brand and a year; the h1 speaks to someone already on the page. They should agree on the same core phrase. This tool measures that agreement as the share of the h1's meaningful words that also appear in the title, ignoring common words, and warns below 30%.
Can this checker read a page built with JavaScript?
Only what the server sends. We read the HTML your server returns and do not execute JavaScript, so a page that renders its headings in the browser will look emptier here than it does to a visitor. That is a real limitation and we would rather say it than quietly report your page as having no h1. If the result surprises you, view the page source rather than the inspector and compare.
What is in the layer that needs an account?
The answer-engine heading layer: whether your h1 is shaped like a question an engine can match to a query, how many of your h2s already are, whether the first two sentences after the h1 actually answer it, and the exact sentence an engine would most likely quote. It is computed from your own page and costs nothing to run. The account exists so we can tell you when the result changes, and because the tools on this shelf that do spend money stay free.
How is this different from other free h1 checkers?
Three things. It reports an empty or image-only h1 as an empty h1 rather than as none, which changes what you go and fix. It gives you the heading outline with the skipped levels named, not just an h1 count. And it shows the threshold behind every warning instead of returning a score you cannot argue with.
More free tools
Every tool on the shelf is free, with no card and no trial. Most need no account at all. See all free tools.
Check whether AI assistants recommend your product for a real buyer question, and see which competitors they name instead.
Generate the buyer prompts AI assistants actually get asked about your category, grouped by funnel stage.
Build a valid llms.txt file for your site, including the citation and safety blocks most generators leave out.
See the web searches an AI engine actually runs before it answers a buyer question, straight from the engine, not a template.