SEO Tools

Heading outline diff.

Paste the old and new HTML of a page. The tool extracts each H1-H6 outline, aligns them with a longest-common-subsequence diff, and shows what was added, removed, moved or re-levelled.

Advertisement

Inputs

Paste raw HTML source. Nothing is uploaded and pasted markup is never loaded or executed.

Result

Paste old and new HTML, then press Compare.

Advertisement

How the heading outline diff works

Each input is parsed for native H1 through H6 elements and elements with role="heading" plus aria-level, in source order. The two lists of (level, text) pairs are aligned with a longest-common-subsequence diff, the same idea used by text diff tools. Text is compared after collapsing whitespace and ignoring letter case.

What each label means

The diff is a structural comparison of heading text and level only. It does not judge whether a change helps or hurts search performance, and duplicate headings are paired in source order.

Sources

MDN heading-element and ARIA guidance reviewed September 30, 2026.

Advertisement

Common mix-ups

  1. W3C WAI: Page structure, headings

Checked against these sources on .

Heading outline diff FAQ

Why compare heading outlines before and after a redesign?

Templates and CMS migrations often reorder, rename or re-level headings without anyone noticing. A diff makes those structural changes visible so you can confirm they were intended.

What is the difference between moved and level changed?

Moved means the heading kept its level and text but sits at a different place in the outline. Level changed means the text is the same but the level (H2, H3 and so on) is different.

Does a changed H1 count hurt rankings?

The tool only reports that the count differs. A single descriptive H1 is the clearest pattern, but this tool makes no claim about ranking effects.

Can it fetch two URLs?

No. Browsers block reliable cross-origin fetching, so paste the HTML source of each version.

Is pasted HTML uploaded?

No. Parsing and comparison run locally in your browser, and pasted markup stays inert so embedded resource URLs are not loaded. The page itself still uses the site's analytics and advertising services.

Does it see headings that JavaScript adds after the page loads?

Only if they are in the HTML you paste. A page's source usually contains the HTML as the server sent it, so for client-rendered headings copy the rendered markup from your browser's developer tools instead and paste that into each box.

Why should a page have only one H1 and no skipped levels?

MDN recommends preferring a single h1 per page and nesting headings without skipping levels, because assistive technology users often navigate by jumping between headings and rely on the levels to understand how sections relate. The tool flags a changed H1 count and any heading more than one level deeper than the one before it.

Why compare outlines at all

Headings are the outline of a page. MDN notes that screen-reader users commonly navigate by jumping from heading to heading to find out what a page contains, and recommends preferring a single <h1> per page and nesting headings without skipping levels. W3C's accessibility tutorial on headings makes the same point: a heading's level says how it relates to the ones around it. When a template change, a CMS migration or a redesign touches the markup, headings are among the first things to be re-levelled, merged or dropped without anyone deciding to. A line-by-line diff of the HTML buries that in noise; a diff of just the outline shows it at once.

A worked example, with the real output

Suppose the old page has this outline:

H1 Trail Shoe Guide
  H2 Fit
  H2 Cushioning
    H3 Midsole
  H2 Price
  H2 FAQ

and the redesigned page has:

H1 Trail Shoe Guide
  H2 Fit
  H2 Price
  H2 Cushioning
  H2 Midsole
    H4 Weight
  H2 FAQ

Pasting both into the tool gives four unchanged headings and three findings:

The H1 count is one on each side, so no H1 warning appears. Each of these changes may be intentional, but none of them would show up in a visual check of the new page, and the skipped level in particular is the sort of defect that survives review.

How to use it in a redesign or migration

  1. Before the change, copy the HTML of each important template (home, category, article, product) and keep it.
  2. After the change, copy the same pages from staging. If headings are built by JavaScript, copy the rendered HTML from your browser's developer tools (Elements panel, copy the outer HTML of the main element) rather than the page source, because the source may not contain them.
  3. Paste old and new, run the diff, and walk through the list. For each removed or re-levelled heading ask whether it was a deliberate editorial decision.
  4. Fix the skipped-level and H1-count flags first; they are the cheapest to correct and the easiest to justify.
  5. Keep a short log of accepted differences so the next release is compared with the new baseline.

Common mistakes

How matching works, and its limits

Headings are matched on level and text after collapsing whitespace and ignoring letter case, so "Hello World" equals "hello world". When the same text appears more than once, matches are paired in source order. Because it works on pasted markup, the tool does not fetch URLs, execute JavaScript, or judge whether a changed outline helps or hurts search performance; it reports structure only. For the tags in the head of the page, use the HTML Head Diff Checker.

Sources

Each source was opened and checked against the statements above in October 2026. Standards and vendor documentation change; follow the links for the current text.