Knowledgebase

Measuring Real User Performance Print

  • frontenddevelopment, frontend, performance, guide, howto, solution, zillionkinghost, hosting
  • 0

Field data.

WHY IT MATTERS

Laboratory measurement is controlled and repeatable. It is not what your visitors experience.

WHAT FIELD DATA REVEALS

The range of devices and connections in use Which pages are slow for real people How performance varies by location

WHAT TO COLLECT

The core timing measurements, from real sessions.

WHAT TO SEGMENT BY

Device type Connection quality Location Page

WHY SEGMENT

Averages conceal that a substantial proportion of visitors have a poor experience.

WHAT TO LOOK AT

The worse end of the distribution, not the average.

WHAT THAT USUALLY SHOWS

Considerably worse figures than laboratory testing suggested.

WHAT TO ALSO COLLECT

Errors occurring in browsers, which you otherwise never learn about.

WHAT TO BE CAREFUL WITH

Collecting more personal data than needed The monitoring itself adding weight

WHAT TO ACT ON

The pages with the most traffic and the worst figures.

WHAT TO MEASURE AFTERWARDS

Whether the change actually helped real visitors.


Was this answer helpful?
Back

Are you happy with your experience? Leave us a review on Trustpilot.


Trustpilot