In This Article
Mobile passed 60 percent of global internet traffic in 2024 and has held above that share through 2025 and into 2026. StatCounter’s running count puts the early figure at 64 percent mobile, 33 percent desktop, and 3 percent tablet. The shift is sharpest in Asia-Pacific (around 70 percent mobile) and least pronounced in North America (around 55 percent), but the direction of travel is consistent everywhere.
This guide covers what the 2026 numbers actually say, why the shift is structural rather than seasonal, and the practical implications for anyone who builds or markets digital products in the back half of the decade.
TL;DR
The pick: Mobile is around 64 percent of global web traffic in early and growing slowly; the desktop versus mobile split is now permanent.
Runner-up: The biggest delta since 2020 is in commerce, where mobile share of transactions has finally caught up with mobile share of sessions.
Skip if: Skip mobile-first design only if you build B2B software for enterprise office users; even then, expect roughly 30 percent of sessions on phones.
What the current numbers actually say
StatCounter, Cloudflare Radar, and Akamai all publish overlapping mobile-versus-desktop traffic figures and the picture is consistent. Global mobile share sits around 64 percent in early, up from 50 percent in 2018 and 58 percent in 2022. The growth rate has slowed to roughly one percentage point per year, which suggests we are near a structural ceiling for mobile share.
The 5 percent that desktop has lost since 2022 went mostly to mobile rather than tablets. Tablets continue to languish around 2 to 3 percent of traffic globally, a category that never quite found its footing the way smartphones did.
Why the shift is structural
Three forces locked the share in. First, smartphone ownership crossed 80 percent in most developed markets by 2024, so the addressable population is functionally everyone with internet access. Second, app-store distribution has reduced the cost of mobile-first publishing to near zero. Third, public Wi-Fi and 5G coverage have made mobile data effectively unlimited for typical browsing in OECD countries, removing the one friction that kept some users on desktop for long sessions.
The pandemic-era bump that pushed people to desktop for remote work has fully unwound. Workers who returned to office or to hybrid setups largely returned to mobile-first browsing on personal time, and the desktop share that briefly recovered in 2020 to 2021 is now firmly back to its 2019 trajectory.
What is different versus 2022
The biggest delta is commerce conversion. Mobile share of e-commerce transactions caught up with mobile share of sessions in 2025 for the first time. Through 2022, mobile drove most of the visits but desktop closed most of the high-value purchases. Today the mobile checkout experience (Apple Pay, Google Pay, one-tap PayPal, native passkeys) has closed the gap. Shopify reports 71 percent of its transaction count and 58 percent of its gross merchandise volume now come from mobile.
The other big delta is search behaviour. Google’s AI Overviews and ChatGPT’s browser plug-ins have shifted high-intent search queries off the traditional results page on mobile faster than on desktop, which has implications for any business that depends on click-through from search.
What the shift means for builders
Mobile-first is no longer a slogan, it is a baseline. Anyone designing a web product should be sketching the mobile breakpoint first, building the desktop layout as an enhancement, and testing critical paths on a real low-end Android device before launch. The mid-tier 2026 Android phone is a 250 USD device with 4 GB of RAM and a slow eMMC drive; if your homepage takes more than three seconds to interactive on that hardware, you are leaving real money on the table.
The exception is B2B SaaS for enterprise office users, where desktop remains primary. Even there, mobile sessions account for around 30 percent of total, mostly for notifications, approvals, and quick lookups. A read-only mobile companion to a desktop-primary tool is the right minimum.
Practical Core Web Vitals targets
Google’s 2026 Core Web Vitals thresholds are LCP under 2.5 seconds, INP under 200ms, CLS under 0.1. On real-world mid-tier Android hardware, hitting LCP under 2.5 seconds requires hard discipline: aggressive image optimization (AVIF or WebP), a sub-100 KB critical CSS budget, and JavaScript shipped incrementally rather than as a single 1 MB bundle.
INP became the dominant interaction metric in 2024 replacing FID. The biggest INP wins come from breaking up long tasks, deferring non-critical hydration, and avoiding synchronous third-party scripts in the critical path. Most sites fail INP because they ship analytics, A/B testing, and consent banners in a way that blocks the main thread during input.
What to actually do this quarter
Audit your top five pages on a real Android phone (Pixel 8a or a 250 USD Motorola), record a session, and time every critical interaction. If anything takes more than two seconds, prioritize fixing it before any new feature work. Use Chrome DevTools’ mobile emulation as a daily check but never as a substitute for real-device testing; emulation underestimates network and CPU constraints in ways that matter.
Build a budget. Set a JavaScript budget per route, set an image weight budget per page, set a third-party script budget for the whole site. Track them in CI. The teams that ship fast websites do it because they treat performance as a budget constraint, not as an aspiration.
How should you weight mobile?
- Consumer brands: Mobile-first, desktop as enhancement. Aim for 80 percent of effort on mobile.
- E-commerce: Mobile checkout is the conversion path. Treat as primary.
- B2B SaaS: Desktop primary, mobile companion for notifications and approvals.
- Local services: Mobile-first; the search and click happen on the phone.
FAQ
Will mobile keep growing?
Slowly, yes. The current 64 percent figure is likely to drift to 67 or 68 percent by 2030 as desktop continues to shed casual browsing. The big growth phase is over.
Should I still build a desktop site?
Yes for B2B, mostly yes for consumer. Desktop sessions are longer and convert at higher values even when fewer in number. Build responsively, prioritize mobile, but do not abandon desktop.
How important is PWA?
Less than expected. Native apps and responsive web have eaten the PWA promise. PWAs still make sense for specific cases (offline-first tools, install-friction-averse audiences) but are not the default.
What about AR and VR?
Mostly noise outside specific verticals. The Vision Pro and Quest 3 install bases are small enough that they round to zero in global traffic figures.
Bottom line
Mobile finished its takeover of internet traffic somewhere around 2024 and the 2026 numbers reflect a settled equilibrium near two-thirds of global sessions. The practical takeaway is operational: build for mid-tier Android first, treat performance as a budget, and design your conversion paths assuming a phone in one hand. Desktop is not dead, but it is no longer the default mental model for anyone designing for the open web.
