What is PLS by connection type
PLS by connection type breaks down your Perceived Load Speed (PLS) by the user’s network connection, such as 4G, 5G, 3G, and others. This view helps you understand how different types of connections affect how fast the page feels to load.
PLS is a real-user performance metric developed by Uxify that captures the time it takes for all meaningful elements on the screen to load, based on what users perceive—not just what the browser sees. Since PLS reflects the real feeling of speed, slower network conditions often have a direct impact on it.
Healthy PLS by connection type sample
Should you worry
A healthy PLS by connection type breakdown usually shows better performance for users on faster network like 5G. While some variation between connection types is expected, major slowdowns (especially on common networks like 4G) should be investigated. If many of your users rely on a slower connection, optimizing for that segment can yield big engagement wins.
Unhealthy PLS by connection type sample
An unhealthy connection type breakdown often reveals that users on slower or unstable networks (e.g., 3G, unknown) are experiencing longer PLS. If these users make up a sizable portion of your audience, they may be leaving before the page ever feels usable — not due to broken functionality, but because it simply feels too slow.
Resolving unhealthy PLS by connection type
Go-to action plan to resolve an unhealthy PLS by connection type:
Ask Uxi to analyze your PLS by connection type and suggest improvements.
Use Filters to identify which pages cause the most slowdown and compare them to other PLS breakdowns.
Simulate LCP of the suspected breakdown to see if fixing it will resolve the PLS by connection type. If yes, this is where the resolution focus should be.
Use an automated optimization tool like Navigation AI to improve your PLS by connection type.
Once you’ve improved PLS, set an alert to be the first to know if it starts worsening again.
Try it yourself
Discover how your website performs with real user data.