When a user’s real journey spans two devices, an organic search touch on a phone, followed later by a conversion on a laptop through a direct visit or a branded paid search click, standard attribution measurement breaks that journey into two entirely separate, unlinked user records unless cross-device identity resolution is explicitly in place. The mobile organic touch gets no credit at all in the eventual conversion’s attribution data, not because organic search didn’t matter to that outcome, but because the measurement system has no mechanism to know the mobile session and the desktop session belonged to the same person, absent a deliberate identity bridge.
Why cross-device journeys break standard attribution measurement
Standard web analytics measurement is built around the browser or device as the unit of identity by default, typically a client identifier stored in a cookie or similar mechanism, tied to a specific browser on a specific device. A user who searches on their phone and later opens a laptop to actually make a purchase is, from the perspective of that default measurement approach, two different unrelated “users,” each with their own separate session history, unless something explicitly connects the two.
GA4 offers a mechanism to bridge exactly this gap: User-ID, which lets you attach your own authenticated user identifier (for instance, a logged-in account ID) to sessions across whatever devices that same identified user happens to use. When a user is logged in on both their phone and their laptop, GA4’s User-ID reporting can, in principle, stitch those two sessions together as belonging to one identified person, allowing the mobile organic touch and the desktop conversion to be recognized as part of the same journey rather than as two disconnected events. Google Signals, when enabled, offers a related but distinct cross-device signal based on users who are signed into their Google Account and have enabled ads personalization, providing another potential bridge for cross-device recognition under Google’s own aggregated reporting.
The critical constraint is that both of these mechanisms depend entirely on the user actually being identifiable and logged in (or otherwise Google Signals-eligible) across both touchpoints. If the user searched on mobile without logging in, then converted on desktop either without logging in or by directly navigating rather than clicking a tracked, identity-linked path, there’s no bridge available, and the two sessions remain measurement-independent even though they were, in reality, the same person completing one continuous journey. This is the mechanism behind the specific distortion in the question: the mobile organic touch is real, and may well have been the actual origin of that user’s eventual decision to convert, but standard attribution has no way to credit it unless identity resolution specifically connected the two device sessions.
The scale of this problem for any given site depends heavily on login rates and Google Signals opt-in rates among its actual users, which varies enormously by site type, industry, and user behavior; there’s no fixed, universal percentage of conversions “lost” to this gap that applies broadly, because it’s entirely a function of how many of a given site’s users are identifiable across devices at all, which is specific to that site’s own audience and product.
How to close the mobile-to-desktop attribution gap
Where realistic given your product (sites with logged-in user accounts, memberships, or authenticated experiences), implementing GA4’s User-ID feature properly, consistently tying the same identifier to the same user across every device and session where they’re authenticated, is the direct mechanism available to at least partially close this gap. Enabling Google Signals, where appropriate for your property and privacy posture, provides an additional cross-device signal layered on top, based on Google’s own signed-in user recognition rather than your own login system.
Beyond implementation, it’s important to set honest expectations about what’s actually measurable here rather than presenting a false sense of completeness. Even with both mechanisms fully implemented, cross-device measurement only captures the portion of your traffic that happens to be identifiable, logged in, or Signals-eligible at both ends of the journey; users who never log in, or who convert through a genuinely anonymous browsing session, remain structurally invisible to cross-device stitching no matter how well the identity-resolution tools are configured. Reporting on cross-device attribution should reflect this as a partial, best-available view rather than implying the mobile-to-desktop gap has been fully solved, since for any site with a meaningful share of non-authenticated traffic, it hasn’t been and can’t be, given the tools currently available.
