Does mobile-first indexing really ignore the desktop version of your site?

Why does Google still sometimes check the desktop version?

Mueller's response is clear on one point: indexing is based on mobile. But the interesting detail is the mention of "relevant links" that are still checked desktop-side.

This means that Google has not completely abandoned desktop, but uses it as a secondary source for validation. In practice, we observe that certain signals like complex internal links or specific navigation structures can still trigger occasional desktop crawling.

What happens if my mobile content differs from desktop?

This is where things get tricky. If your mobile version hides entire sections via accordions, collapsed tabs, or worse, removes content altogether to "lighten up", Google only sees what is rendered on mobile.

The old "mobile light" techniques where you removed 50% of the text to gain speed? SEO suicide today. Your mobile content must be comprehensive, even if the display is optimized differently.

How does Google handle sites without a mobile version?

If your site doesn't have a mobile or responsive version, Google still indexes something — but it's your desktop version viewed through a mobile Googlebot. It's safe to say the experience is disastrous and you will lose positions.

It's not a question of "if" but "how much" you are losing. Field data shows drops of 20% to 60% in organic traffic on non-responsive sites since the complete shift to mobile-first indexing.

• The mobile version is now the primary index — the desktop is only a one-off verification source for certain signals
• Any content absent from mobile is invisible to Google, even if it exists on desktop
• Desktop checks mainly concern links and certain complex navigation structures
• A non-responsive site is indexed through forced mobile crawling, with disastrous ranking consequences
• Strict mobile-desktop equivalence has become a prerequisite, not an option

Is this statement consistent with field observations?

Yes, in essence. Since the full deployment of mobile-first indexing, it’s indeed observed that content missing from mobile disappears from the index. A/B testing where text is strategically removed from mobile shows almost immediate drops in positions.

However, the mention of "relevant links" remains vague. Google does not specify which links trigger a desktop check, nor how frequently. On sites with complex architectures (e-commerce with thousands of categories, media with complex taxonomies), we sometimes observe sporadic desktop crawls without a clear pattern. [To verify]

What nuances should be added to this rule?

The devil is in the implementation details. Content being present does not mean it's accessible. If your mobile text is technically in the DOM but hidden behind 3 clicks or aggressive lazy-loading, Google may see it but assign it a token weight.

Another point rarely mentioned: images and media must also be equivalent. A site loading desktop images in 2000px and mobile versions in 400px with different alt texts creates inconsistency. Google indexes attributes from the mobile version — if your mobile alt texts are poor, you lose semantic signal.

When does this rule still pose problems?

Complex B2B sites and SaaS platforms are often collateral victims. They have historically favored the desktop experience with rich interfaces (data visualization, comparison tables, configurators). Adapting all this to mobile without losing functionality is a real puzzle.

The result: they either keep the mobile content lightweight and lose SEO, or they force all desktop content onto mobile and face Core Web Vitals penalties. A vicious cycle that Google never openly acknowledges.

What should you prioritize checking on your mobile site?

Your first reflex: crawl your site with a mobile user-agent. Screaming Frog, OnCrawl, or even the Search Console with the "smartphone" filter will show you what Google actually sees. Then compare with a desktop crawl — any difference in content, internal links, or meta-data is a red flag.

Next, manually test your strategic pages on real mobile devices (not just the responsive mode of Chrome). Does the content display without user action? Are accordions deployable but closed by default? Do tabs hide key text? Google reads the full DOM, but user experience indirectly impacts ranking.

What critical mistakes should definitely be avoided?

Don’t fall into the trap of "simplified mobile". Removing sections of content under the pretext of mobile UX is suicidal. If a block of text seems too long, rework the layout, use expandable snippets, but don’t delete anything.

Another common mistake: images lazy-loaded too aggressively. If Google has to virtually scroll 5 times to load a strategic image, it may never be seen. Native loading="lazy" attributes are fine, but poorly coded custom JS scripts create black holes in your indexing.

How can you ensure that Google indexes the right version?

The Search Console remains your source of truth. Check the "Coverage" tab with the smartphone filter active — all your critical pages should be indexed in mobile version. If you see important pages marked as "Detected, not indexed" in mobile but indexed in desktop, that’s a warning sign.

For sites with separate versions (mobile.*, m.*), ensure that each desktop page has its properly annotated mobile equivalent. Just one broken canonical link and it’s Russian roulette between the two versions. Homogeneity of annotations is critical, not optional.

• Crawl the site with a mobile user-agent and compare it line by line with the desktop crawl
• Ensure that 100% of the desktop textual content is present and accessible on mobile
• Test images: same alt texts, same relative dimensions, guaranteed loading
• Check that mobile internal links are equivalent to desktop (same linking, same anchors)
• Validate canonical/alternate annotations if mobile/desktop versions are separate
• Monitor the Search Console: all strategic pages must be indexed in smartphone version