The mobile version leaves out internal links desktop has

Why it matters

Google indexes the mobile version of a page, and it uses links to find pages to crawl and to judge their relevance; it can only follow a link written as an <a> element with an href.

What to do

Review the desktop and smartphone counts, total missing count, missing-target samples and measured variation in the finding.
Keep relevant links in the smartphone source; a menu may be collapsed with CSS while its links remain in the HTML.
Run another site audit after the correction. A fix requires a complete clean comparison of the same requested page and final address.

When to leave it as it is: if the missing links are desktop-only extras and every page they point to is linked elsewhere on the mobile site, you can leave it; the check does not measure how important the links are.

What we read

We compare a frozen sample of representative pages admitted in this audit. The sample grows with the audit page count, from 10 to 100 when enough eligible pages exist, and is bounded by the available eligible pages. The home page and unresolved same-subject findings have priority, followed by pages from different site sections. The report shows the actual sample size and page count.

There are four observations: smartphone Chrome and desktop Chrome, each confirmed twice with the audit identity. The first smartphone observation reuses the primary crawl response; one smartphone confirmation and two desktop fetches are shared across the mobile comparisons. We read retained HTML without executing JavaScript.

We compare distinct normalized internal targets from links in the HTML. Both corresponding desktop/smartphone pairs must miss more than 30% of desktop targets or at least 20 targets, and the missing fraction must exceed measured same-role variation by more than 0.1. Same-role variation above 0.1, a desktop without links or incomplete link evidence leaves the result unknown. A script shell is not automatically excluded.

Example & evidence

Illustrative example — one homepage on desktop and on phones, not a live audit.

Address: https://bakery.example/. Desktop Chrome gets 40 distinct internal links: 8 in the menu, 12 in the main content and 20 in the footer. Smartphone Chrome gets the menu and main content without the footer in both observations:

<footer>
  <!-- 20 links on desktop, none in the phone HTML -->
</footer>

Half the desktop targets are missing on smartphone in both comparisons, with stable repeated sets, so the page is reported with missing-target samples.

A possible correction: keep the 20 footer links in the phone HTML; they can be collapsed with CSS. Moving them into a <template> for a script to copy would still be reported.

Confirm & check again

Check for yourself: view the page source as a phone, for example with your browser's device mode, and compare its footer and menu links with the desktop source.

Check again: Full audit — A later audit fixes the finding only after a complete clean comparison of the same requested page and its native final-address identity. An unread or unselected comparison, a changed final address or a different representative carrier does not prove a fix. Dismissal does not remove the technical recheck.

Limits & sources

Only retained HTML targets are compared; JavaScript execution, CSS visibility and link importance are not assessed. The sample is not the whole site.

Robots denial, challenge responses, missing or incomplete observations, unstable links and a desktop without links remain coverage limits. A complete own comparison can still be usable when another sampled page is unread.

Confidence: Medium because the comparison is repeated on served HTML, without rendering or a link-importance assessment.

Sources

Google, Mobile site and mobile-first indexing best practices — says Google uses the mobile version of a site's content for indexing and ranking. Checked 25 September 2026.
Google, Link best practices for Google — says Google uses links to find pages to crawl and can only crawl an <a> element with an href. Checked 25 September 2026.

Editorial review: 4 November 2026

The mobile version has less or different content — compares the main content's words instead of the links.
The mobile version drops or changes headings — compares the headings instead.

The mobile version leaves out internal links found on desktop