The Device dimension shows the type of device on which the user searched: Mobile, Desktop, or Tablet. Comparing Mobile and Desktop can reveal where reported visibility or click behavior differs, but the raw totals are rarely a fair contest because search demand and user behavior vary by device.
The most useful question is not “Which device has more clicks?” It is “For the same scope, which pages and query families behave materially differently, and is the gap in impressions, CTR, or average position?”
Device data is search context, not page format
A Mobile row means the search occurred on a mobile-class device. It does not mean:
- Google indexed a separate mobile URL.
- The user saw the same result layout as desktop.
- Mobile and desktop had equal query demand.
- A mobile performance gap was caused by page speed.
- Google used a different canonical URL for every device.
Google uses the mobile version of a site’s content for indexing and ranking under mobile-first indexing. That crawling/indexing policy is different from the Device dimension, which segments performance by the searcher’s device.
Create the comparison in Search Console
- Open Performance → Search results.
- Select the relevant Search type and a complete date range.
- Add a Device filter.
- Choose Compare.
- Select Mobile and Desktop.
- Turn on Clicks, Impressions, CTR, and Average position.
- Open Pages, then Queries.
- Add a Country or Page filter if the audience or site area must be held constant.
Search Console allows one comparison at a time. If Device is the comparison, compare dates in a separate saved/exported view or export both periods for downstream analysis.
Use Tablet as a separate segment when it has enough volume to matter. Do not automatically combine it with Mobile; Google reports it as its own device category.
Control the scope before interpreting the gap
Device mix varies strongly by:
- Query intent.
- Country.
- Search type.
- Page template.
- Brand versus non-brand demand.
- Weekday and time-sensitive behavior.
- Search appearance and result layout.
For a fair diagnosis, hold as many of these constant as possible. “Mobile CTR is 2% and desktop CTR is 5%” at property level can simply mean mobile has more impressions for broad, lower-position queries.
A better sequence is:
Country → site section → page → query family → device comparison
Use a Domain property when separate m. and www. hosts must be included in one broad view. Otherwise, separate URL-prefix properties can hide half of a legacy mobile/desktop configuration.
Diagnose the four metrics
Mobile clicks lower, mobile impressions lower
Possible explanations:
- The topic has lower mobile search demand.
- Important mobile query families are missing.
- Mobile visibility is concentrated in fewer countries.
- Mobile content or metadata parity problems reduce relevance.
- Separate mobile URLs have crawling, redirect, canonical, or indexing errors.
Search Console alone cannot separate market demand from lost opportunity. Compare the same page/query combinations and investigate technical parity.
Mobile impressions similar, mobile CTR lower
Investigate:
- The same queries on both devices.
- Average position within those queries.
- Mobile result layouts and search features.
- Whether titles truncate or communicate poorly on a smaller screen.
- Whether a rich-result appearance differs.
- Local, app, image, or other result features competing for attention.
- Whether intent becomes more immediate on mobile.
Do not assume a title rewrite is the answer until position and query mix are controlled.
Mobile average position worse
Filter to an affected page, then inspect queries. A property-level position difference can be a mix effect: Mobile may expose the site to a larger number of lower-position queries.
For recurring high-impression queries with a sustained gap, inspect:
- Content equivalence between mobile and desktop rendering.
- Robots meta tags and blocked resources.
- Structured data parity.
- Title and meta-description parity.
- Internal links visible in rendered mobile HTML.
- Mobile redirects and error behavior.
- Intrusive ads or content hidden behind required interaction.
Google recommends equivalent primary content, meaningful headings, metadata, and structured data across mobile and desktop versions. Important content can be placed in accordions; it should not require an interaction before Google can load it.
Mobile impressions higher, clicks not proportionally higher
This often reflects broad mobile demand, lower positions, different query intent, or result-layout competition. Prioritize pages where the gap remains after controlling for query and country and where impression volume makes improvement material.
Mobile clicks higher despite lower CTR
Higher mobile demand can produce more clicks even with weaker CTR. Do not describe this as simple underperformance. Report both the scale and efficiency perspectives.
Page-level workflow
- Compare Mobile versus Desktop.
- Open Pages.
- Export if you need calculated gaps or minimum thresholds.
- Identify pages with meaningful impressions on both devices.
- Rank by lost-click opportunity or CTR percentage-point difference.
- Select one page and open Queries.
- Compare the same visible queries.
- Inspect the page on mobile and run technical/experience diagnostics.
Useful exported fields include:
| Field | Why it matters |
|---|---|
| Mobile/Desktop clicks | Outcome scale |
| Mobile/Desktop impressions | Demand and visibility context |
| CTR difference in percentage points | Click-behavior gap |
| Average-position difference | Supporting visibility evidence |
| Minimum impressions | Removes volatile small samples |
| Country and query family | Controls mix |
| Template | Finds systematic mobile issues |
Estimate click opportunity carefully
A directional estimate can prioritize review:
mobile click opportunity = mobile impressions × (reference CTR - mobile CTR)
The reference CTR should come from comparable queries or pages at similar positions, not automatically from the entire desktop site. Desktop CTR may be unattainable on mobile because the result layout and intent differ.
Label this as an estimate, never guaranteed traffic. Exclude tiny samples and cap negative opportunity at zero for prioritization.
Query-level workflow
For each important page:
- Find queries with material impressions on both devices.
- Compare click difference, CTR, and average position.
- Group close variants into a query family after export.
- Inspect the live mobile and desktop result context from an appropriate location, understanding that results are personalized and time-sensitive.
- Review whether the landing page satisfies mobile intent quickly and completely.
If mobile visibility comes from a different query set, improve or create content only when those queries align with the site’s purpose. More impressions are not automatically desirable.
When to investigate Core Web Vitals
A device gap is a reason to check experience data, not proof that speed caused the Search gap.
Move to the Core Web Vitals report when:
- The affected URLs belong to templates with Poor or Needs improvement mobile groups.
- Mobile landing behavior in GA4 is also weak.
- Rendering or interaction problems are reproducible.
- A release coincides with both experience degradation and Search change.
- The gap is concentrated on pages sharing front-end components.
Search Console’s Core Web Vitals report uses real-world Chrome User Experience Report data and separates Mobile and Desktop. It groups similar URLs and assigns a group status based on its worst-performing metric. It is not designed as a lookup for every individual URL, and insufficient-traffic URLs can be absent.
Core Web Vitals are part of page experience, but Google does not say that improving a score guarantees a ranking increase. Fix experience problems because they affect users and may align with what ranking systems seek to reward; preserve the distinction between evidence and causation.
Responsive sites versus separate mobile URLs
Responsive design
The same URL and HTML normally simplify canonicalization and parity. Investigate CSS, layout, rendering, interaction, and device-specific result behavior when a gap remains.
Dynamic serving
The URL is the same but HTML varies by user agent. Verify the Vary: User-Agent response and compare rendered content, metadata, structured data, robots directives, images, and links.
Separate m. URLs
Check:
- Equivalent mobile URLs for each desktop page.
- Correct redirects by device.
- Canonical from mobile to desktop and appropriate alternate relationships.
- Consistent status codes.
- No mobile
noindexor robots block. - Equivalent structured data and metadata.
- Both hosts included in the property scope.
Multiple desktop pages redirecting mobile users to one mobile homepage is a serious configuration problem, not a CTR optimization task.
Aggregation and canonical caveats
The chart is property-aggregated, while the Pages table is page-aggregated. Table totals can differ from the chart when several pages appear for the same query.
Most performance data is assigned to Google’s selected canonical URL. Google also notes that a device-specific duplicate may sometimes be better suited to a user even when another URL is canonical. For separate mobile URLs, review canonical and device configuration before interpreting URL-level winners and losers.
Hidden queries and top-row truncation also mean visible mobile and desktop query rows may not reconstruct total clicks.
What Search Console can answer
It can answer:
- How reported clicks, impressions, CTR, and position differ by device.
- Which pages and visible queries show the largest gaps.
- Whether a gap is concentrated by country, Search type, or appearance.
- When a device-specific trend began.
It cannot prove:
- That page speed caused a ranking or CTR gap.
- That mobile and desktop users had the same intent.
- How the page behaved after the click.
- Which exact result features competitors occupied.
- Whether a user switched devices.
- Revenue or conversion impact.
Use GA4, CrUX/PageSpeed Insights, rendering tests, server logs, and result inspection for those questions.
Review checklist
- Device is treated as searcher context, not page format.
- Country, page, query, and Search type are controlled where practical.
- Both absolute volume and rates are reported.
- Average position is used as supporting evidence.
- Pages with meaningful volume on both devices are prioritized.
- Mobile content and technical parity are checked.
- Core Web Vitals are investigated without assuming causation.
- Canonical and aggregation rules are respected.
- Small and hidden-query samples are disclosed.