Yes, it is ok. Google has said plainly, several times over several years, that a page can rank with one H1, five H1s, or none at all, and the HTML specification permits any number. That is the direct answer, and you can stop reading if rankings were the whole question. The more useful question is whether you should, and there the honest advice has not changed since this post first ran in 2019: use one H1 anyway. The reasons have nothing to do with Google and everything to do with the people and machines that read your page structure, which is why the full story is worth ten minutes. It runs through what Google actually said, a browser feature that never existed, and a screen reader shortcut most developers have never watched anyone use.
The short version
- Google has no penalty and no preference. Multiple H1s do not hurt rankings.
- The HTML5 promise that made multiple H1s respectable was never implemented by any browser and left the spec in 2022.
- Screen reader users navigate by headings, and one clear H1 orients them instantly.
- Use one H1 on pages you control. Do not launch a project to fix templates that have two.
What Google has said on the record
The clearest statement came from John Mueller in a 2019 episode of the Ask Google Webmasters video series, answering this exact question. His answer, paraphrased: the site will rank perfectly fine with no H1 tags or with five of them, Google's systems are not that picky, and headings help Google understand sections of a page rather than acting as a ranking switch. He has repeated the position in Search Central office hours many times since, often sounding mildly baffled that the question keeps coming back. Nothing in Google's documentation contradicts him. The Search Central guidance on headings talks about using them to convey structure and hierarchy, and never sets a count.
The reason the question keeps coming back was captured neatly by SEO researcher Cyrus Shepard, who polled SEOs on Twitter and found a solid majority believed Google itself recommends a single H1. Google never said that. The industry inherited the rule from somewhere else and then misremembered where, and the actual origin story explains a lot about whether the rule deserves to survive.
Where the one-H1 rule came from
In the early 2000s, search engines leaned hard on markup clues to work out what a page was about, and the H1 was the loudest clue on the page. One H1 carrying the target phrase was measurable, teachable advice, and it hardened into doctrine through a decade of audit checklists. Modern Google reads the content itself far better than it once did, so the heading count stopped being a lever. What headings still do, and Mueller made this point in the same video, is describe what each section is about, which feeds how Google maps a page, pulls featured snippets, and matches long-tail queries to specific passages. That work gets done by heading quality, not heading count.
The HTML5 outline algorithm made it respectable, then never shipped
There was a period when multiple H1s were not just tolerated but officially encouraged. HTML5 introduced sectioning elements, article, section, aside, and nav, along with a document outline algorithm that was supposed to compute heading levels from nesting depth. Under that model you could open every section with an H1 and the browser would understand one as a top-level heading and another as effectively an H3, based on how deeply it sat. Tutorials taught the pattern. Validators accepted it. Boilerplates shipped it.
No browser ever implemented the algorithm. Chrome, Firefox, and Safari all continued to treat an H1 as an H1 regardless of nesting, and so did every screen reader and every search engine, meaning pages built on the pattern had flat, misleading outlines the whole time. After years of the spec describing behavior that did not exist anywhere, the WHATWG removed the outline algorithm from the HTML standard in 2022, and the spec's advice now points the other way: use one H1 per page and lower heading levels for subsections. The strongest technical argument for multiple H1s died the day it was supposed to have been born, and it took a decade for the paperwork to catch up.
The accessibility case for a single H1
This is where the recommendation stops being a style preference. Screen reader users move through unfamiliar pages by jumping between headings, pulling up a list of them, or keying straight to the H1 to learn what the page is about. WebAIM's screen reader surveys have found, year after year, that heading navigation is the most common way users explore a new page, ahead of reading straight through or searching within the page. For those visitors the heading structure is not decoration, it is the map.
One H1 makes the map instant: jump to it, and you know where you are, the same orientation a sighted visitor gets from the big visual headline. Three H1s force a different experience, announcing three top-level topics on a page that presumably has one, and leaving the listener to work out which heading is the real title and which are a logo, a tagline, and a section header that someone styled as H1 because the font size looked right. WCAG does not literally require a single H1, but the accessibility community recommends it almost unanimously, and MDN's guidance says the same. When a rule costs nothing and helps the people who depend on structure most, the ranking question stops being the interesting one.
The audit call we make in practice: heading problems are worth fixing when they mislead, not when they miscount. A page whose three H1s are the logo, the page title, and a newsletter box has a clarity problem worth cleaning up on the next template touch. A page with two honest H1s from an old HTML5 theme is fine, and rebuilding it for that reason alone is money spent on a problem no user and no crawler actually has.
Why your audit tool keeps flagging it
Part of the reason the debate refuses to die is that the software keeps reopening it. Most site auditors, including the audit tools in Semrush, Ahrefs, and Sitebulb, flag multiple H1s or a missing H1 as a warning on every crawl, and a warning in a report reads like a problem to whoever receives the report. The tools are not wrong to check, since a clean single H1 is the convention they can verify at scale, but the severity gets inflated in the retelling. Treat the flag the way the tools themselves usually label it, as a notice about convention rather than an error, and prioritize it below anything that touches indexing, redirects, or page speed. An audit that leads with heading-count warnings is telling you the crawler found nothing worse, which is good news dressed up as a finding.
How to check a page in under a minute
You do not need a crawler to answer the question for any single page. Open the browser console and run one line:
document.querySelectorAll('h1') The result lists every H1 on the page, and clicking each entry highlights it in the document, which is how the surprise logo H1 usually gets discovered. For the fuller picture, the free HeadingsMap browser extension draws the whole heading tree the way a screen reader user would experience it, skipped levels and all. Reading your own page as that outline, stripped of fonts and layout, is the fastest way to feel whether the structure communicates or confuses, and it takes less time than the average debate about heading counts.
When multiple H1s are genuinely fine
Some real situations produce more than one H1, and knowing which ones matter keeps the fix list honest. A theme that wraps the site logo in an H1 on every page, alongside the article's own H1, is the single most common case we crawl into, a leftover convention from old WordPress themes. Pages assembled from independent components, embedded widgets, or syndicated fragments sometimes carry an H1 they did not choose. And single-page applications that stack several views into one document can end up with an H1 per view. In every one of these cases the page ranks on its content, screen readers cope, and the correct response is to consolidate whenever the template is open for other reasons, not to schedule emergency work.
The heading structure we still recommend
One H1 stating what the page is about, matching the intent of the title without needing to copy it word for word. H2s for the major sections, H3s inside them, levels never skipped on the way down, and heading tags never chosen for their font size, because CSS exists for that. Write each heading as a plain description of its section, since that is the part Google actually uses, and let keywords appear where the description naturally produces them. The full breakdown of how the six levels work together, including the skipped-level and styling mistakes that flatten page outlines, is in our post on header tags. Heading structure is also one of the cheap checks inside a wider technical SEO audit, where it usually takes minutes to review and years to have gone unnoticed.
Questions people ask about multiple H1 tags
Does Google penalize multiple H1 tags?
No. There is no penalty and no ranking requirement either way. John Mueller of Google has said on camera that a site ranks fine with several H1s or with none at all, and he has repeated the point in office hours sessions for years.
Why do SEOs still recommend a single H1?
Because one H1 keeps the main topic of the page unambiguous for machines and people, gives screen reader users a single clear answer to what the page is about, and costs nothing to do. The recommendation survives on clarity grounds, not ranking ones.
Is multiple H1s valid HTML?
Yes. The HTML specification permits any number of H1 elements. What changed is the advice around it: after the outline algorithm was dropped from the spec in 2022, the spec itself began steering authors toward one H1 per page with lower levels for subsections.
Should I rebuild my template to remove a second H1?
On its own, no. If a redesign or template change is happening anyway, consolidate to one H1 while you are in there. As a standalone project it will not move rankings, and the development time is better spent on content or real technical problems.
Do keywords in the H1 still matter?
Relevance matters more than exact matching. Google reads headings to understand what a page and its sections cover, so the H1 should describe the page accurately. If the keyword belongs in that description naturally, include it. Forcing it in changes nothing worth having.