This translation has not been editorially reviewed yet. The German version is authoritative. Deutsche Fassung →
Measuring mobile and desktop separately: why a good value says nothing about the other
Mobile and desktop are treated as two separate measurements, because they run under different conditions: throttled connection, weaker processing power, smaller viewport. A good desktop value says nothing about the phone.
- The difference is usually substantial, and not a measurement error. The same amount of JavaScript costs many times more time on a mid-range phone.
- Because most traffic happens on mobile, the mobile measurement is usually the one that matters — even if your own staff only ever look at the page at their desk.
FAQ
Why is the mobile value so much worse?
Because it's deliberately measured with a throttled connection and lower processing power — that reflects reality outside fast office networks.
Should we optimise both values?
The mobile one first. What helps there almost always helps on desktop too; the reverse is rarely true.