Mobile-first indexing is Google’s practice of using the mobile version of a page, as fetched by its smartphone crawler, as the primary basis for indexing and ranking. Content, links, structured data and metadata that exist only on the desktop version are, in effect, invisible to Google Search.
In technical SEO, the term describes a change of perspective rather than a separate index. There is one index for all devices, but it is built from what a mobile user agent receives. Desktop searchers therefore also see rankings that are based on the mobile page.
How mobile-first indexing works
Mobile-first indexing works by making Googlebot Smartphone the primary crawler for a site. It requests pages with a mobile user agent and a mobile viewport, renders them, and passes the result to indexing. Googlebot Desktop still visits some URLs, but its fetches no longer determine what is indexed for a site that has been switched.
The rollout took several years. Google announced the approach in November 2016, began moving sites in March 2018, made it the default for new websites from July 2019, and later declared the migration complete. Since July 2024, Google crawls and indexes the remaining sites with its smartphone crawler, and content that is not accessible on a mobile device is no longer indexed.
| Misconception | Reality |
|---|---|
| There is a separate mobile index | One index exists, built from the mobile view of each page |
| It is a ranking boost for mobile sites | It changes which version is evaluated, not the ranking formula |
| Desktop-only content still counts | Content absent from the mobile page is not indexed |
| Content in tabs or accordions is devalued | Hidden but loaded mobile content is indexed normally |
| Mobile-friendliness and mobile-first are the same | Usability is a page experience topic; mobile-first is about indexing |
Mobile configurations and their risks
The site configuration determines how hard it is to keep mobile and desktop in parity. Google supports three configurations and recommends responsive web design because it removes most parity risks by design.
| Configuration | How it works | Main mobile-first risk |
|---|---|---|
| Responsive web design | Same URL and HTML; CSS adapts the layout | Elements removed from the DOM by script on small screens |
| Dynamic serving | Same URL; server returns different HTML by user agent | Missing Vary: User-Agent header or stripped content |
| Separate mobile URLs | Distinct mobile host, often an m-dot subdomain | Broken annotations, thin pages and incorrect redirects |
Separate mobile URLs are the most fragile option. Each desktop page needs a rel=alternate annotation with a media query pointing to its mobile counterpart, and each mobile page needs a canonical annotation back to the desktop URL. Hreflang annotations on an m-dot site must point to the mobile versions of the alternates, and redirects must send each mobile visitor to the equivalent page, never to the mobile home page.
What must match between mobile and desktop
The central requirement of mobile-first indexing is content and signal parity: everything that matters for search on the desktop page must also be present on the mobile page. A layout may differ; the indexable substance must not.
| Element | What to verify on mobile | Risk if missing |
|---|---|---|
| Primary content | Same text, headings and media | Lost relevance and rankings |
| Structured data | Same markup with mobile URLs where relevant | Loss of rich result eligibility |
| Title and meta description | Equivalent wording | Weaker or rewritten search snippets |
| Robots meta tags | No noindex or nofollow added on mobile | Pages dropped from the index |
| Internal links | Navigation and contextual links present | Weaker discovery and link equity flow |
| Images and video | Same high-quality files with alt text | Lower image and video search visibility |
| Hreflang and canonicals | Correct targets for the mobile version | Wrong language or URL selected |
Internal linking deserves special attention. Mobile designs often simplify menus, drop footer links or replace related-content blocks with nothing. Each removed link reduces how well Google discovers and weights deeper pages, which can quietly weaken category and product pages after a redesign.
Rendering and loading pitfalls
Many mobile-first problems are rendering problems rather than design choices, and they overlap closely with JavaScript SEO. The smartphone crawler renders pages in a headless Chromium browser but does not scroll, click or type like a person.
- Interaction-triggered lazy loading, such as a tap to load more, leaves content unloaded for the crawler; loading on viewport entry is safe.
- Blocking CSS or JavaScript in robots.txt prevents Google from rendering the mobile layout and its content.
- Infinite scroll without paginated URLs hides items that load only after scrolling.
- Mobile-only error pages or interstitials served only to mobile user agents can replace real content in the index.
- Slower mobile servers or heavier mobile pages can reduce crawl capacity, which matters on large sites.
Google retired its standalone Mobile-Friendly Test and the Search Console mobile usability report in December 2023. The URL inspection tool, which shows the rendered HTML and a screenshot from Googlebot Smartphone, is now the most direct way to confirm what Google indexes.
Auditing a site for mobile-first readiness
A mobile-first audit compares the two versions systematically rather than relying on visual checks. Experienced teams use the decision logic below to prioritize work.
| Finding | Recommended action | Reason |
|---|---|---|
| Mobile word count far below desktop | Restore the missing content on mobile | Only mobile content is indexed |
| Structured data only on desktop | Deploy identical markup on mobile | Rich results depend on the mobile page |
| Separate m-dot site with frequent errors | Plan a migration to responsive design | Removes duplicate maintenance and annotations |
| Content behind tap-to-load buttons | Render it in the initial HTML or on scroll | The crawler does not click |
| Different robots directives by device | Unify the directives immediately | A mobile noindex removes the page entirely |
A practical method is to crawl the site twice with an audit crawler, once with a desktop and once with a smartphone user agent, and then compare word counts, links, titles, canonicals and structured data per URL. Differences above a few percent on templates that matter deserve investigation.
Stripped to the essentials
Most people search on their phones. So Google decided to judge every website by its phone version. When Google reads a page, it reads it the way a smartphone sees it.
Imagine an extra paragraph or a table that appears only on the computer version. Google may simply not know it exists. The safe approach is simple: the phone version should contain everything important. It may look different on a small screen, but nothing that matters should be missing.
Where it is applied
A home goods retailer ran a separate m-dot site built years earlier. Desktop category pages listed 120 products with buying guides, while the mobile versions showed only 40 products, dropped the guides and omitted the Product structured data present on desktop product pages. After its switch to mobile-first indexing, the retailer saw product rich results disappear and category rankings slip for competitive queries.
The SEO and development teams compared both versions with two audit crawls and confirmed the gaps. As a first step they added the missing structured data and buying guides to the mobile templates; over the following quarter they migrated to a responsive design and redirected every m-dot URL to its equivalent main URL with a 301.
After the migration, Google indexed one version of each page with full content and markup. Product rich results returned within weeks, and category pages recovered their previous positions once the complete product lists were crawlable on mobile.