0

I've been trying for the last couple of days to improve the performance of our Wordpress site for mobile usage. In order to validate the optimisation process, I've been using PageSpeed Insight and Lighthouse (mobile).

PageSpeed Insight Results

As you can see in the first picture, everything was green, but we were at the 2.5s limit for the LCP, so we wanted to optimise more. In order to quickly test various tweaks, I switched to Lighthouse in my browser (incognito mode, no extensions).

Lighthouse Results

You can see the results in the second picture. The results are similar, but a bit more discouraging. So I wanted to check the Original Trace of this Lighthouse run to understand what is leading to 3.0s loading time for LCP.

Original Performance Trace

As you can see in the third picture, surprisingly, the LCP seems to load at about 470-500ms. That is at least 5 times faster than what the Lighthouse benchmark reports. Moreover, the entire page seems to load in less than 1s.

So my question is where is this difference coming from? Why are Lighthouse and PageSpeed Insight reporting different values than the actual Performance trace.

Thank you!

Vlad Stefanescu
  • 9
  • 1
  • [this answer](https://stackoverflow.com/a/64792982/2702894) I gave probably answers your question and also details how to make the performance trace match Lighthouse numbers. – GrahamTheDev Dec 21 '20 at 11:55
  • @GrahamRitchie your answer is exactly what I was looking for! Great explanation! Do you know if PageSpeed Insight is using the same throttling parameters? – Vlad Stefanescu Dec 21 '20 at 14:50
  • Yes they are (very nearly) the same, slight differences due to hardware config, distance to server etc but the difference should be hardly noticeable. – GrahamTheDev Dec 21 '20 at 15:13
  • Good to know! Thanks! – Vlad Stefanescu Dec 22 '20 at 08:32

0 Answers0