Switched on Wordpress APO, yet to see improvements. What am I doing wrong?

Switch on wordpress APO for my blog: https://www.alittlebitofspice.com/ on October 4th.

Saw an almost immediate decrease in server response time:

but its going back up.

And the page load time is back to pre-APO days.

Am I doing this wrong?

APO is working perfectly on your site.

I think so, but I don’t see page speed improvement in the metrics so far.

Shouldn’t be the case. The pages haven’t been updated in 4-5 months nor new pages added.

As I can see, you are using Google Ads and Google Tags scripts. I believe that you can’t achieve significant performance improvements with these scripts.


Google Analytics scripts are ubiquitous, right?

Make sure you’re looking at the right metrics. For WP APO the direct benefit of a cached HTML page is in terms of TTFB - which for Google Analytics closest metric is server response time.

For Webpagetest TTFB is the TTFB of the Index request specifically and NOT Webpagetest reported summary First Byte which is a different metric https://community.centminmod.com/threads/cloudflare-automatic-platform-optimization-for-wordpress-cache-effectiveness.20494/

The First Byte summary is a the total for Discovered + DNS + Connect + SSL + TTFB times. The First Byte is the time taken from navigationStart to the browser receiving the very first byte of data from the server. See Matt Hobb’s explanation here.

First Byte summary not same as Index request TTFB

Lastly, be aware of your visitor traffic profile. CF CDN cache has greater benefit for visitors further away from your origin’s geographical location but less of benefit for visitors closest to your origin location. So if 99% of your visitors are in New York and your origin server is in New York, you may not notice a difference in TTFB or server response time. But if your visitors are further away, you will notice.

Google Analytics can break down page speed per country and city locations so you can further inspect average server response time by region. You can also import Google Analytics as a source data connector to Google Data Studio to do more custom dashboards with custom analytic charts/tables to visualize your page speed.

example of my GA stats in Google Data Studio breaking down pages speed by country

and then filtering for server response times by country

USA only

versus Australia

1 Like

GA stats are real world visitors so actual averages depend on visitors connection speed etc. The visitor’s ISP connection speed isn’t constant as are the devices used to connect to your site. Slower mobile/tablet/laptop devices vs fast desktop devices will impact your page load speed metrics reported by Google Analytics.

Also traffic geographic makeup/profile will impact averages. You could have one day 90/10 percentage of close origin visitor versus further away from origin visitors - in which case average reported metrics will be fast. Next day you could have 50/50 percentage which will lower your averages. Then one day you could have 20% from very very slow device visitors - I’ve seen some South African/Brazil visitors on mobile take 60-100 seconds to load my page while in USA visitor loads same page in <1.5 seconds. Those slow visitors will pull down and lower your overall averages. So breaking down page speed metrics by country/region is needed to see a better profile of your page speed traffic profile.

As does what your visitors are doing on their devices concurrently while visiting your web site. Example I posted at Performance Tutorials - Google PageSpeed & Webpagetest.org - #4 by eva2000 how LCP metric can blow out from 2 seconds to 19 seconds by just having a visitor loading my forum page while watching a video on same laptop device which impacts local laptop’s resource and network usage impacting browser page render speeds

This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.