This translation has not been editorially reviewed yet. The German version is authoritative. Deutsche Fassung →
Responsive
What screen sizes should a website be tested on?
A website should be tested at three screen sizes: smartphone at 390, tablet at 768, and desktop at 1440 pixels wide. These three profiles cover the common device classes. The audit service emulates real devices for this — with touch input, a mobile browser identifier, and matching pixel density — instead of just shrinking a browser window.
Three profiles cover the device classes
The smartphone profile measures 390 by 844 pixels — the size of a common current phone. The tablet profile measures 768 by 1024 pixels, a tablet in portrait orientation. The desktop profile measures 1440 by 900 pixels, a common monitor size. At these three widths, most websites change their appearance: the menu collapses, columns stack, images scale. That's exactly why a single glance at your own screen isn't enough — a page can look flawless on desktop and scroll sideways on mobile, cut off text, or shrink buttons. That's why every profile is checked independently and stated separately in the report, along with a screenshot.
Real device emulation delivers real results
A merely shrunk browser window behaves differently from a phone. Many websites only switch to their mobile display once the browser identifies itself as a mobile device: with touch input, a mobile browser identifier, and the higher pixel density of a phone screen. The audit service therefore sets up a full device emulation for every profile — the smartphone profile identifies as a phone with touch and triple pixel density, the tablet profile as a tablet. Without this emulation, false findings would arise: the check would see an in-between form of the page that no real visitor ever sees. With it, it sees the same rendering as the audience.
The check sees the page the way a visitor does
Before measuring, the check lets the page fully load and settle. It then scrolls slowly once to the bottom of the page and back up — the way a visitor would look around. There's a technical reason for this: many pages only reveal content while scrolling. If the measurement happened beforehand, these blocks would still be in their starting position, and the check would report a layout that doesn't exist in real use. Only after this pass does the actual measurement begin: scroll behaviour, controls, text wrapping, content on the first screen — per profile, with a screenshot as evidence.
- Check new pages on at least one real smartphone, in addition to looking at your work monitor.
- Compare the three screenshots in the audit report — they show how the same page looks at every device class.
- Ask your agency at what screen widths your website changes its layout, and have exactly those transitions tested.
- Treat the mobile view as your primary view, since many visitors see your website exclusively on mobile.
FAQ
Is it enough to shrink the browser window?
A shrunk window shows only part of the truth. Real phones identify themselves with touch input, a mobile browser identifier, and higher pixel density — many websites only switch to their mobile display then. The audit service therefore emulates a full device per profile, and thereby sees the same rendering as a real visitor on mobile.
Why is every device size checked individually?
Because websites change their layout with screen width: menus collapse, columns stack, images scale. A finding can occur at exactly one device size — a table fits on desktop and overhangs the edge on mobile. That's why the report states every measurement separately per profile, each with its own screenshot.
What screen width does the test use for smartphones?
The smartphone profile uses 390 pixels wide and 844 pixels tall, at triple pixel density — the size of a common current phone. The browser calculates in so-called CSS pixels; the screen's physical resolution is higher. In addition, the service checks a tablet profile at 768 and a desktop profile at 1440 pixels wide.