Mobile Rankings Tanked But Desktop Is Fine

Published Read 5 min read

On this page

A mobile-only ranking drop has fewer possible causes than it first seems, because Google builds one index from the mobile version of your pages and serves it to every device. If your site is responsive, the mobile and desktop versions carry the same content, so a content problem would show up on both devices. What’s left to check includes whether the split is real, what differs on the mobile results page and in the mobile experience itself, and, if you serve separate mobile URLs or different HTML to phones, whether the mobile version is missing something the desktop version has.

One index, built from mobile

Google announced in October 2023 that the move to mobile-first indexing was complete. In the same post it noted that crawling and indexing as a smartphone meant a mobile page now needed to be as complete as the corresponding desktop version. The version that earns your rankings is the mobile one, whatever share of your traffic comes from desktop.

That has a direct consequence for diagnosis. Google’s mobile-first indexing guide says that with responsive design, the content and the metadata are the same on the mobile and desktop versions. On a responsive site, then, thin content, a lost link or a missing section is a problem for both devices, not a reason for mobile alone to fall.

Confirm the split first

Open the Search Console Performance report and use the Devices tab, then compare the same pages and queries on mobile and desktop over the same dates. Two details from Google’s help page shape how you read it:

  • Average position is the average position of the topmost result from your site.
  • Results differ by device. Google says search results are specific to the time, place, device and recent history of the person searching.

A third-party rank tracker checking mobile and desktop from one location is a sample, not your traffic. If Search Console shows mobile clicks and impressions falling on pages where desktop holds, the split is real. If only a tracker shows it, check that first.

What can differ by device

None of these is proof on its own; each is a difference you can check.

The results page itself. Because results are specific to the device, a change in what Google shows on phones for your queries can move your mobile position with no change on your site, so search your main queries on a phone and compare what ranks around you.

The mobile page experience. web.dev’s Core Web Vitals page sets the 75th percentile of page loads, segmented across mobile and desktop devices, as the threshold to measure, and Search Console’s Core Web Vitals report shows the two separately. The same HTML can pass on desktop and fail on phones: a slower CPU and network, and a narrower viewport that can change which element is the largest contentful paint. Oversized images that CSS shrinks on screen still download at full size, so check which element is LCP on mobile before optimizing.

Dialogs that cover the page. Google’s guide to avoiding intrusive interstitials says that, unless they’re legally mandatory, you shouldn’t obscure the entire page with interstitials. A newsletter or app-install dialog that sits in a corner on desktop can fill the whole screen on a phone. Check each dialog at phone width.

A separate mobile version. If you serve separate mobile URLs, or different HTML to phones on the same URL, the mobile version is its own build and can drift. The mobile-first indexing guide asks you to make sure the mobile site contains the same content as the desktop site, and says having the same content on both versions ensures the two versions can rank for the same keywords. It also asks for the same structured data and metadata on both. A mobile template that dropped a module, a FAQ block or a product description to stay light is a parity gap in exactly the version Google indexes.

Content that waits for a tap

One parity trap looks mobile-specific but isn’t. The mobile-first indexing guide says Google won’t load content that requires user interactions, such as swiping, clicking or typing, to load. Content that is in the HTML but collapsed behind a tab or accordion can be indexed; content fetched only after a tap isn’t loaded. Because the index is shared, content that never loads in the mobile version is missing for desktop results too. If a section like that disappeared, expect both devices to feel it.

Test in layers

Each layer can catch problems the previous one misses:

  1. DevTools device emulation for layout, viewport and whether content renders at narrow widths. It doesn’t reproduce a phone’s CPU.
  2. Lighthouse in mobile mode for lab performance under simulated mobile conditions and for the LCP element.
  3. A real mid-tier phone on a cellular connection, where CPU cost and image weight show up.
  4. URL Inspection’s live test to see the page Google fetches, with the Crawled as field showing the user agent type used for the test, and to compare its HTML with the desktop page.
  5. Analytics by device. If engagement or conversion falls on mobile while desktop holds, it points to where the experience breaks.

Start with the mobile Core Web Vitals and the LCP element, then dialogs, then parity if you run a separate mobile version.

Frequently asked questions

My traffic is mostly desktop. Does the mobile version still decide my rankings?

Yes. Google indexes from the mobile version of your pages and serves that index to every device. A desktop-heavy audience doesn’t change which version Google indexes.

Could a mobile-only drop be a content problem?

On a responsive site, content is the same on both versions and a content problem would show on both devices. With separate mobile URLs or different HTML for phones, a content gap in the mobile version is possible, and Google says the same content on both versions is what lets them rank for the same keywords.

How long does recovery take after the fix?

Google has to recrawl the fixed pages, and field data in PageSpeed Insights covers the previous 28 days, so improvements show up gradually rather than on the day you ship.

Related posts:

  1. How to Leverage Log File Analysis for SEO Decisions Server logs are the fullest record of what crawlers requested from your site, when they include your edge traffic: every URL, response code, timestamp and user agent. Search Console tells......
  2. How to Do SEO During a Website Redesign A redesign can lose traffic through defects you can check for before and after launch, such as URLs that changed without a redirect, pages removed while they still earned links,......
  3. Hiring SEO Talent: Interview Questions and Skill Assessment Methods SEO hiring is hard for a structural reason: the field has no standard credential, and organic results are shaped by enough factors that a candidate can sincerely claim growth they......
  4. SSR vs CSR vs ISR: Rendering Strategy Selection for SEO For SEO, the rendering choice turns first on one question: is the page's content in the HTML the server sends, or does it appear only after JavaScript runs in a......
  5. Why Query Parameters After Hash Symbols Don’t Exist to Google Everything after the # in a URL is a fragment, and the browser doesn't send it to the server. So when a filter lives after the hash, /properties#bedrooms=3 and /properties#city=seattle......
  6. SEO Process Documentation: Playbooks, SOPs, and Knowledge Management Documentation earns its keep when following it is cheaper and safer than doing the work from memory. Building for completeness can work against that test. Pages nobody opens still need......
  7. How to Reverse Engineer Google’s Quality Raters Guidelines for Your Site Google's Search Quality Rater Guidelines are a public description of what Google means by a helpful, trustworthy result, which makes them a practical checklist for auditing your own site. Use......
  8. Google Discovers Your Site But Won’t Index It When Search Console lists pages as "Discovered - currently not indexed", their content has not been evaluated and rejected. Google hasn't crawled them yet. Google's Page indexing report help defines......
  9. How to Do SEO for Directory Websites A directory has a case for ranking when its curation visibly adds something a searcher can't get from Google directly. Google already shows local businesses with their own profiles, maps......
  10. How to Do Image SEO Google does not index images set as CSS backgrounds. Its image SEO documentation says so in one line, and the rule applies before any question of alt text: a hero......
  11. How to Do SEO When Rebranding (Same Domain) A rebrand that keeps the domain leaves the URLs, content and links in place, so non-branded rankings have no rebrand-specific reason to move. The exposure sits wherever Google shows or......
  12. How to Do SEO for YouTube Videos Two people who type the same search on YouTube can see different results. YouTube's explanation of how its search works says its system may consider a viewer's search and watch......
  13. How to Do SEO for Event Pages Event-page SEO can fail when each event is treated as disposable: a fresh URL every year, optimized in the run-up to the date, then abandoned. That can throw away what......
  14. How to Diagnose and Fix Crawl Traps A crawl trap is a pattern on your site that generates an endless supply of low-value URLs: a calendar that pages forward forever, a session ID added to every link,......
  15. How to Build a Multilingual SEO Strategy The strategic decisions come before any hreflang tag. Multilingual SEO is a sequence of investment choices: which languages to build, whether to target a language or a specific country, which......

Leave a comment

Your email address will not be published. Required fields are marked *