Semantic HTML means choosing the element that matches the meaning of the content: an h2 for a section heading, a ul for a list, a table for tabular data, a nav for navigation, a main for the main content and an a element with an href for a link. Screen readers depend on it, and it makes the structure of a page plain to any program that parses the HTML, including search crawlers and the systems that pull passages into AI answers. Google does not describe semantic elements as a ranking boost, but some basics, such as real links and real headings, affect whether content is found and understood at all.
Need help with SEO?
Comppound runs SEO and AI search for SaaS companies. We plan what to publish, produce it with AI, and every page passes automated checks, and a person reviews samples and anything high-risk.

- Published to your site and tracked in Search Console
- One person answers for your account
- Plans from $400 a month, terms from 3 months
The elements that matter most
- a with an href for links, which is what Google follows; click handlers on spans are not crawlable links
- h1 to h6 in order, so headings describe the outline of the page
- ul, ol and li for lists, which keep steps and options as separate items
- table with th headers for comparisons and pricing data
- main, nav, header, footer and article to separate the content from page chrome
- img with alt text, and button for actions
<!-- Hard to parse -->
<div class="title">Plans</div>
<div class="row"><div>Starter</div><div>Free</div></div>
<!-- Clear to people, screen readers and crawlers -->
<h2>Plans</h2>
<table>
<tr><th>Plan</th><th>Price</th></tr>
<tr><td>Starter</td><td>Free</td></tr>
</table>Why it matters for AI search
Retrieval systems break pages into passages. A comparison in a real table, steps in an ordered list, and a heading that names the question all survive that process with their meaning intact. The same content built from styled divs can lose its structure, which makes it harder to quote accurately. This is a practical argument from how parsing works, not a documented ranking rule.
Example
A pricing table rebuilt
A fictional payroll tool, Wagecraft, builds its plan comparison from nested divs with CSS grid. Rebuilding it as a table with header cells changes nothing visually, makes it readable to screen readers, and gives crawlers a clear mapping of plan to feature.
Common mistakes
Choosing headings by font size
Pick the heading level for its place in the outline, then style it. Skipping levels for looks confuses the structure.
Buttons that navigate and links that act
Use a for going somewhere and button for doing something. Crawlers follow links, not buttons.
Frequently Asked Questions
Google does not say so. It helps crawlers find links and understand structure, and it is required for good accessibility.
They help assistive technology and keep markup clear, but real links, headings, lists and tables do most of the work.
Keep reading
- Header Tags: Header tags (H1 to H6) are the HTML headings that structure a page. They help readers scan and help search engines understand how the content is organized.
- Web Accessibility: Web accessibility means building sites that people with disabilities can perceive, understand, navigate and use, including with assistive technology.
- Extractability: Extractability is how easily a machine can pull accurate, self-contained facts and passages out of a page, a key trait for content used in AI answers.
- Content Chunking: Content chunking is writing and structuring a page in clear, self-contained sections so that people and AI systems can use each part on its own.
- Auto-Generated Content: Auto-generated content is text produced programmatically. It violates Google's spam policies when made at scale mainly to manipulate rankings.
- Boilerplate Content: Boilerplate content is text repeated across many pages, such as navigation, footers and stock paragraphs. Too much makes pages look alike.
- Cloaking: Cloaking is showing search engines different content from what users see, to manipulate rankings and mislead people. Google treats it as spam.
- Content Brief: A content brief is the plan for a single page: the search it targets, who it is for, what it must cover and how it should lead readers to the product.
