My blog went from having only 2 green TTFBs to achieving green on all 10 TTFBs (I know, not an official test, but that’s what I got). I got excellent results with APO in Ke圜DN. Just remember when it’s time to purge cache and retest, give it a minute to purge before retesting. Make sure you test your website before and after APO. That’s why I recommend a before & after test in Ke圜DN which measures TTFB and other metrics in 10 global locations. Sure, APO may be able to also improve core web vitals like LCP (and feel free to benchmark those too) – but those tests won’t give you the full picture. The whole point of APO is to cache HTML pages to Cloudflare’s worldwide data centers, which means you should see the biggest improvement in locations far away from the origin server. That’s because they only test your site in 1 location. Tools like GTmetrix and PageSpeed Insights aren’t great for testing APO performance. Purge cache in your cache plugin, then Cloudflare.Create an API token and add it to the plugin.Install the Cloudflare WordPress plugin.APO FAQs (answered by Cloudflare, published by Freelancer Tools).APO misconceptions (by Brian Li at Kinsta).I’ve already dug through these and included the recommendations, but feel free to read them yourself (there’s a lot of good stuff): Here are a few helpful resources I used while writing this. Of course, there are a few prerequisites and “small details” which I’ll be covering in this guide. Setting up Cloudflare’s APO on WordPress is easy: once you’ve purchased APO, install the Cloudflare plugin, create an API token in your Cloudflare’s dashboard, and add it to the plugin. While Cloudflare and Kinsta ran their own benchmarks, APO also made a big improvement for my own site which you can see in Ke圜DN. And since WordPress sites are dynamic, Cloudflare is able to serve your entire site from their edge network instead of only static content. Most if not all CDNs (including Cloudflare) don’t serve HTML from the CDN – but APO does. This is due to longer distances between your server and testing location/visitors. When testing your site in Ke圜DN’s performance test, you’ll probably notice how TTFB gets slower as your website is tested further away from the origin server. And yes I have caching plugins installed which minify but the woocommerce stores have some pages that are excluded from caching and this does seem to include exclusion from minification of even the HTML.Cloudflare APO is one of the easiest ways to reduce TTFB in multiple global locations and is easy to set up on WordPress. The whole site minification looks to be of a greater benefit. Should I disable the Railgun and go back to using their nameservers for the minification benefit? I think that actually Railgun isn't worth it for my websites, because they aren't being updated dynamically that often - small woocommerce stores, a coppermine gallery and blogs. I guess, the minification service really only works when you are using the Cloudflare nameservers? Is there something else that I'm missing? The information on both sites is kind of vague on what is working and when. So, I'm using A2hosting which give Railgun for free but you have to setup your DNS records with a CNAME configuration from their cPanel, use the A2hosting nameservers and use their Cloudflare app.īefore that I was using the nameservers which Cloudflare game me. I'm not even sure if this is the right place for this question. Then it got me by reading something remotely connected to my problem somewhere around here or serverfault/stackoverflow that it's related to the Railgun configuration. First I thought it was related to caching. I have burned the last two days wondering why the minification from Cloudflare doesn't work and what is happening.
0 Comments
Leave a Reply. |