Latest news of the rocketat the speed of light!

Why you Shouldn’t Care About Google PageSpeed Insights

As a website owner, you know your site needs to be fast. You’ve read all the articles about how to make WordPress faster and which plugins to install to accomplish this. You’ve probably added a caching plugin, hopefully WP Rocket, and now you want to know how much benefit you’re getting.

So you head over to Google PageSpeed Insights, because that’s what all the articles tell you to do, and enter your URL. You’ll be presented with a grade and a list of recommendations from Google and at that point, you might be dismayed:

  • “What are all these red and orange warnings??”
  • “Why isn’t my grade higher??”
  • “What do all these recommendations mean???”

After adding caching to your site,  you might be expecting that your PageSpeed grade will be near-perfect. Or you’ll look at the recommendations and wonder why your caching plugin hasn’t fixed them all, automatically.

A lot of customers ask us why their PageSpeed grade isn’t higher, or  they assume that because it didn’t increase a lot,  it must mean WP Rocket isn’t making their site faster.

The simple truth is this:

Your Google PageSpeed score does not matter.

That’s right, I said it doesn’t matter. 

Here’s why.

The Need for Speed

The purpose of WP Rocket is to make your site faster.
Speed, i.e the loading time of your site is the most important metric. This is what counts for user experience and for SEO. When the Google bot crawls your site, it cannot see your “grade”, only your speed.

Did you know that Google PageSpeed doesn’t even measure the loading time of your site?

Read that again:

Google PageSpeed does not actually measure the speed of your site. Click To Tweet

We like to use Pingdom Tools to measure the load time:
How To Correctly Measure Your Website’s Page Load Time

Think back to your school days. Did perfect grades mean you were smart? Not necessarily. It just meant that you knew how to do well on tests. But many intelligent people simply not do well on tests.

So just like school grades are not an indicator of intelligence, Google’s PageSpeed grade is not actually an indicator of speed.

Here are 3 websites all with similar load times, but with vastly differing PageSpeed scores:

pureplanet
PurePlanetRecycling
Loading time: 462 ms
Google PageSpeed: 91
anticatrattoria
AnticaTrattoriaAlPortico.it
Loading time: 596 ms
Google PageSpeed: 58
cheeky-punter
CheekyPunter.com
Loading time: 495 ms
Google PageSpeed: 77

Between these 3 sites, the loading time ranges from 465 ms – 596ms but the PageSpeed scores range from 58 – 91.

And the below site has a really good PageSpeed score but is slower than all 3 above:

clnet
Clnet-solution.my
Loading time: 3.16s
Google PageSpeed: 85

So you can see from these examples that the Google PageSpeed grade is not an indicator of speed.

Google's PageSpeed grade is not an indicator of speed. Click To Tweet

Chasing A Grade Is A Waste of Time

No site gets a perfect grade, in fact it’s pretty much impossible to achieve, and since it doesn’t correlate to speed, why bother?

If you try an attain a perfect grade, by implementing all the suggestions Google PageSpeed makes, you will lose your sanity pretty quickly.

You cannot take too literally all of the suggestions from Google PageSpeed because sometimes they are unrealistic or impossible.
For example, it may tell you to minify or add expiry headers to a file that is not hosted on your website. This is impossible. In the below example, only Facebook and Google themselves can add browser caching to these files.

PageSpeed_Insights  - browser caching

If you use a CDN on your site, PageSpeed might give you a lower score for that, but in many cases a CDN provides better speed for your international visitors.
Or if it is marking you down because you could save 1kb by compressing an image (see below), it’s not worth bothering about!

Google PageSpeed Insights Optimize Images

Render-blocking Javascript and CSS

A common recommendation PageSpeed likes to makes is to:
“Eliminate render-blocking JavaScript and CSS in above-the-fold content”

It typically then provides 2 parts to this recommendation:

1) Remove render-blocking JavaScript

It’s preferable for performance that JavaScript files are loaded in the footer of your site, or asynchronously so that they don’t block the downloading of other assets on your site, therefore slowing it down. But what Google does not take into consideration is that sometimes it’s not possible to load everything in the footer, especially if you are not a developer and you are using an off-the-shelf WordPress theme.

If you have activated JS minification in WP Rocket, as of v2.6 we auto-detect whether JS files should be loaded in the header or the footer, based on how their developer enqueued them.

So if you are seeing the “remove render-blocking JavaScript” notice, it means that some JS files are still being loaded in the header. But this may have been intentional on the part of the theme or plugin developer. Sometimes when you move JavaScript files from the header to the footer, you can actually break functionality on your website.

So, if you want to improve your PageSpeed score you can try to use this option in the Advanced options of WP Rocket:
“JS files to be included in the footer during the minification process” and enter the original, full urls of your JS files.

Or if you are not minifying the files but just want to defer the loading you can use the option:
“JS files with deferred loading:”

If this doesn’t break any functionality on your site, it may help your PageSpeed score.
Also remember to test the effect on the speed of your site.

2) Optimize CSS Delivery of the following

WP Rocket does not have an option to load CSS in the footer, because it would cause your site to load without any styles, making it look broken and providing a terrible user experience.

Google suggests you can “Optimize CSS delivery“.

Essentially they want you to separate out the CSS needed to render the first part of your page and embed it directly into the code of your page, rather than in the main stylesheet.

So as you can gather, if you are not a developer, this is pretty hard to do and involves re-coding parts of your theme. It will also vary from site to site, so it’s not something that can be made into a one-click solution in a caching plugin.

If you are a developer, or have one at your disposal you can ask them to implement this technique. It won’t necessarily improve the overall load time of your page, but it will increase your PageSpeed score, and may appear to load faster because the elements at the top of the page will load sooner. This type of micro-optimization is utilized by sites like Amazon where a 10ms difference may impact their bottom line. For most small sites, it’s unlikely to have much impact.

So What Is PageSpeed Good For?

Google PageSpeed can be helpful as long as you don’t treat it as the be-all, end-all.

Sometimes it can alert you to problem areas on your site that you can address. For example, it might alert you to the fact that your content is not being GZIP-ed. WP Rocket adds the rules for GZIP by default so if PageSpeed gives you a warning, this might be a sign your server does not have it activated.

Or it might alert you that you have too many large images which could be compressed. This is a good recommendation which you can actually act on:
Image Optimization – An Easy Win For A Faster Site

We are going to launch a new image compression service – you can sign up for early access now:
https://imagify.io/

Below is a great example of an opportunity to improve because you can save 750kb, an 87% reduction in size. This would have a good effect on your loading time:

optimize-images-large-savings

So it’s best to look at PageSpeed as one of several tools in your arsenal that might provide some pointers, but your goal should always be to improve your actual speed, not your “PageSpeed” grade.

Guidelines for Using PageSpeed

  • Do not blindly trust Google PageSpeed or take it at face value.
  • Always read the recommendations carefully and assess if they are possible and worth your time. If it’s asking you to do something impossible, you should ignore that!
  • Don’t forget to always focus on speed and don’t worry about chasing a grade.
  • Always use an actual speed testing tool like Pingdom to see the impact of any changes you have made on your site.

About

Lucy is a long time user of WordPress, (since 2004) and has provided WordPress training for bloggers and businesses for over 5 years. She has spoken at several WordCamps and is an active member of the WordPress community. You can follow @webtw on Twitter.

Subscribe to our newsletter Latest news of the rocket, at the speed of light!

111 comments

Comment

Another Great Post,
I am using Pingdom Tools and GT Metrix for testing my website.

“No site gets a perfect grade, in fact it’s pretty much impossible to achieve, and since it doesn’t correlate to speed, why bother?”

Check out – http://fastwp.de
…then say it again 😛

But i agree… pagespeed have a couple of good recommendations, but in the end too much “bullshit” and sometimes the recommendations are not for every project a good choice. Be carefull when you trust all this things from google and catch the score. When you not know what you doing, some of thesere tricks and recommendations slow down your site more and don’t came with a speed up.

Thanks, This is good to know! I have spent far to much time chasing the pagespeed grade.

For all our Assessments we use a whole bunch of speed testers, Google Pagespeed is not even one of them 🙂
Great article, keep ’em coming!

This article is for me 😀 Thanks lucy 😉
I think the best record of gtmetrix score is belonged to sahifa demo page. 99 for google and 94 for yahoo!
themes.tielabs.com/sahifa/

I find this post to be wrong on some points. Now I will admit that Page Speed Insights is not the best tool and some things that they recommend can’t be solved its just how things are. However, you are wrong on some several key points.

1. The Optimize CSS delivery. This actually greatly improves the load time of the website and won’t cause the “unstyled” content that you mention. You later seem to correct the point but it is going to confuse users.

What the optimize CSS delivery means, is to load the critical css in the header with an inline style. This will load let’s say in some cases 10kb of CSS and the rest will be loaded with async in the footer. This greatly reduces the load time as the main css to load the page is there to avoid that “unstyled content” and the rest of the extra styling (some widgets or other less important code) will be loaded afterwards. This is great for mobile devices where the download speed can be really poor,

On mobile devices the user will be able to read the content and interact with the site a little bit before the rest of it loads thus greatly improving the overall load time and the time it takes for the user to have a readable website.

2. ALL JS should be loaded in the footer. This is the standard and it should be pushed for every website. Even JQuery can be added to the footer so it should be a priority to build your site with this in mind. While WordPress Rocket doesn’t have a “one click” solution you need to push for it to improve the load time of your website.

Now I agree that things like the save 3 bytes in your JS file is a waste of time. Or compress your image to save 1kb of space is not worth your time.

But the general rule of thumb is anything on your server (caching, headers, etc, etc) or JavaScript and CSS files you should optimize.

WP Rocket Support Team

Hey Scott,

1. The CSS position doesn’t change at all the global loadtime.
This will change a little bit the perception. Could be good indeed to have 2 CSS like Lucy says in the post and what you explain (that’s basically the same).
That’s impossible to set up automatically in a WordPress website.

Putting everything in the footer is terrible for user experience. Here is a great exemple : http://recordit.co/Rh7uZPsFU1
With a slower connexion that’s even worst.

So our recommendation is : it’s better to have everything in the header compare to everything in the footer.

And the best recommendation is to have a small CSS file and to load them conditionally.

2. Nope, nope and nope 🙂 But to moderate my words web performance rules can’t be general and are different website by website.

Do you know that iOS won’t display ANYTHING until it finds the JS files.
So putting JS in the footer will be absolutely terrible for the critical rendering path.
We can see it on your own website : http://www.webpagetest.org/result/150618_WW_D2Q/1/details/
You will see that the start render (the green bar) starts when all the JS are found (not loaded).

Plus regarding to user experience, most of the websites have JS interactions on the top (sliders for exemple), so loading jQuery and all dependencies at the end could be bad. This means that all the interactions are not working and even sometimes not displaying until everything is loaded.

So in my opinion the rule is the same than CSS. jQuery and what we see directly on the header, and the rest on the footer.

Very nice post, thanks for this info

I agree wit the article content but some clients want to see a 100/100 on Goggle page speed insight.

To explain the meaning of a site speed i Wrote this article http://www.azzali.net/il-vostro-sito-e-abbastanza-veloce-per-google/(sorry Italian only) where at the end i put a table with speed results after installing WP rocket cache plugin and obviosly a mention of it.

WP Rocket Support Team

@Fabrizio
That’s your job as a professional to explain to your clients that chasing the 100 is worthless, like we do with that blog post.
I agree, that’s very difficult to do.

@Jean

It’s true that there is no way to do it automatically in WordPress but that does not mean it can be ignored. Your main goal is to get the content visible to the reader as quickly as possible. By optimizing the load of CSS you will see a decrease in bounce rate. Which is the main goal of the webmaster, more views and a lower bounce rate.

If that’s what Safari mobile does than its breaking the standard and need to rectify the mistake. Safari also occupies an extremely small percentage of the market and does not dictate how your website should be optimized. The standard and recommendation is for all JS to be in the footer and to be loaded with async. Therefore what I said was right Safari is at fault here and needs to fix their browser which is already one of the worst on the market.

It doesn’t matter if some websites use a slider on my own websites I use a slider and I can get all of my JS deferred in the footer and its still functional. Again a website built properly can get it to work and be optimized for speed.

That aside its a matter of fact here. Large CSS in the header slows down the rendering and is a pain for mobile users. JS should be in the footer and loaded with async while we inline the critical CSS and load the rest with async.

WP Rocket Support Team

@scott

Remember this blog post in focused on our market which is WordPress.

“Safari also occupies an extremely small percentage”.

Q4 2014: Chrome 37%, IE 14%, Safari iOS 12%, Firefox 11%.
That’s not what I call an extremely small percentage.

Regarding JS, the webperformance community is rather agree that JS should be split in header/footer and not only footer.
Plus you can notice that all the biggest websites (FB, Instagram, Airbnb, etc) are doing this (header and footer). I’m pretty confident that these guys have an infinity more experience compare to you or me.
So we can say there are no debate on this point 🙂

Some statistics say different (more or less for Safari).

That being said most users are not using safari therefore we need to accommodate for the majority of users.

On the JS its decides it goes in the footer. And the only time it should be in the head is if there is a dependency or something that you simply can’t solve in which case you still have performance degradation.

It’s true that Facebook and other companies are doing it but they are running on top of the tier hardware and have the bandwidth to support faster download speeds. While most of our setups are shared hosting vps or rarely dedicated / multiple dedicated servers. So it truly comes down to their infrastructure at that point in which case even though they are not doing the standard they still load very quickly lol.

I think what your plugin does is perfect. It makes websites fast for those who other wise would have no clue what they are doing lol. Your plugin makes WordPress simple because users don’t have to play around with W3 or Super Cache for hours with settings they don’t know how to use.

I just think we should put more focus on educating people what the standards are and even if your plugin doesn’t offer said feature you compare the amount of time invested vs the overall improvement.

the google speed should be only reference to see where your website is not soo good.. but it is wery tricky sometimes

A great and informative article filled with – “useful time-saving facts”!
Much appreciated! Sharing this one.

TWEET THIS BOX
By the way….. could you please point me in the direction on how to create the nifty “tweet this” box function you have in the article? Loved it.

Keep up the good work,
-Erik

So, overall I like the intent of the article, but you’re spreading some wrong information. Google is not perfect and neither are their tools, but if you really want lots more organic traffic from showing up high in Google search results pages, then you need to pay attention to what Google’s tools share with you.

I was recently at a SES convention in Miami where the head of Google Product Development was speaking and she shared (regarding mobile optimizing your site) that even if your site is perfect, but not in the eyes of Google, it will not show up appropriately. It has to be “Google Mobile-Friendly”. In this case, your sites need to be fast and load properly in the eyes of Google, and the insights that they provide are in most cases the same indicators they use to serve up the best user-experience for visitors.

It is tough to make a website get an “A” in Google Page Speed Insights, but it is possible and worth it. They are creating and using established Best Practices that all web developers should be using. Until some other site can provide you the same amount of traffic or more, then we need to pay attention to the tools they provide.

It’s preferable for performance that JavaScript files are loaded in the footer of your site, or asynchronously so that they don’t block the downloading of other assets on your site, therefore slowing it down.

There are a few other items about JavaScript in the footer in the comments as well.

If you look at how current versions of Chrome behave though, you’ll see that just putting JavaScript in the footer doesn’t always change things.

You can use this page as an example. The “devicepx-jetpack.js” reference is at the bottom of the page, but it is the 11th resource to be loaded ( http://www.webpagetest.org/result/150624_MG_1A5F/1/details/ ) out of 51. The pre-loader scans the HTML looking for script references, then gives them a higher priority in requesting them.

I’ve written about this previously https://josephscott.org/archives/2015/04/scripts-at-the-bottom-is-not-enough/

Hello,

This plugin below, ensures PageSpeed 100/100
https://wordpress.org/plugins/speed-booster-pack/

My question is:
 
This plugin is compatible with wp-rocket ?

Regarding Google PageSpeed agree on everything 100%. It makes sense to look to his rekkomedatsiyam but PageSpeed not an indicator of speed. But if the option Rocket lazy loading images, the Pingdom does not have time to upload pictures, so the speed is much faster. Actually option lazy loading of images – this is very good in terms of speed and does not interfere with the reader.

I understand what you are trying to say and agree with most conclusions here, however I have some doubts about using Pingdom’s Load Time as the main indicator of site’s speed. It’s a common agreement that what matter most to users is a perceived performance.

For example check http://www.smashingmagazine.com/ with cleared cache, as a WP site with 99/100 in Google PageSpeed test. It feels like displaying instantaneously due to prioritizing above the fold content and resulting speed index below 1000. And yet the load time in the Pingdom test is above 2 seconds.

http://tools.pingdom.com/fpt/#!/cxWU85/http://www.smashingmagazine.com/
http://www.webpagetest.org/result/150716_Q9_MT3/

Not saying Speed Index should be the only performance indicator, for example additional network activity can slow down browsers especially on mobiles but should be at least mentioned here shortly.

Keep up great job.

WP Rocket Support Team

@Lubos

I partially agree with your point. It’s important to properly measure the load time. Smashing Magazine is using a CDN, to when you are using PingDom it’s very important to do multiple tests to be sure that the static content is properly cached by the CDN pop servers.
But yes

For exemple after a few tests, 797ms => http://tools.pingdom.com/fpt/#!/bQNg0e/http://www.smashingmagazine.com/

In the other hand, I fully agree that Smahing Magazine have done great optimization regarding to perception of the load time 🙂

I am using Wp Rocket after several other plugins and I am surprised for the good control of the source code and the results in term of pure speed.

My question is about this point: “When the Google bot crawls your site, it cannot see your “grade”, only your speed”.

Are you totally sure that Google bot doesn’t use any check system alike the score parameter using in Google Speed Test?

WP Rocket Support Team

@Enrico: I think it doesn’t make senses for the Google bots to make a Google Speed test for these reasons:
– they have to be fast to crawl a website, a pagespeed test takes is very long
– you can have 100% and a page loading time to 10 seconds, so Google Bot should give you some “points” only for the grade? No, it’s why the only metrics that everybody must take care is the page loading time

that’s a great tool and very powerful plugin and helpful support
my site: http://alafdal.org/
load speed was 16 s or more
after installing the plugin and activate – huge difference happened
you can check now my speed 654ms

that’s the great number ، that’s a dream didn’t come in my mind

A good advace how we achieved 100 Insight Score for our web site http://www.itransition.com – We set up an additional check for user-uploaded images – some of them were not completely optimized and had unnecessary metadata etc. The module performed lossless-optimization for these images.

You may be right, that Google does not really measure the correct speed, but the way Google measures speed is the way Google used it in its SEO-Relevant rankings. So Google Page speed still is important.

Thanks for these insights! I will send clients here when they complain about their score.

It makes me chuckle when Google suggests that I fix one of their problems:

Minify JavaScript:
Minifying https://www.gstatic.com/…ha/api2/r20150928175454/recaptcha__en.js could save 555B (1% reduction) after compression.

or

Optimize images:
Losslessly compressing https://maps.gstatic.com/…mapfiles/api-3/images/tmapctrl4_hdpi.png could save 1,000B (22% reduction).

@ Johnny
Show to your client the result of Page Speed Tool on some Google Pages.
As you can see below are not so brilliant.

In Italy we say “il calzolaio che va in giro con le scarpe rotte”. Translated in English language sound like the “The cobbler who walks with broken shoes”.

Result in page speed for Google News Italian Page: Mobile speed 73/100 – Mobile User Experience 94/100
https://developers.google.com/speed/pagespeed/insights/?url=https%3A%2F%2Fnews.google.it%2Fnews%3Fcf%3Dall%26hl%3Dit%26ned%3Dit

or

Result in page speed for Youtube Home page: Mobile speed 61/100 – Mobile User Experience 100/100
https://developers.google.com/speed/pagespeed/insights/?url=https%3A%2F%2Fwww.youtube.com%2F

or

Result in page speed for Google Developper Page: Mobile speed 65/100 – Mobile User Experience 100/100
https://developers.google.com/speed/pagespeed/insights/?url=https%3A%2F%2Fdevelopers.google.com%2F

you are not right. I got for https://meko.lv/ 100/100 for first page and otpimised most landingapges. This is new website and it is ranking fast . If you get 100 your page will be faster just because less http requests and smaller code.

This article so wrong on the functional analysis of page speed insights. It may be correct in saying it doesn’t matter [to the average wordpress blog owner]. But is perfect page speed necessary for the average wordpress blog? No. But for developers working on “real” business websites (lets call then “enterprise” websites), this article is completely misleading and it may be worth it to gain a perfect score across all these tools.

Additionally a number of your statements are simply factually false. PageSpeed Insights DOES measure your page speed the same way Pingdom does. It will warn you if the markup takes more than 200ms to download and reduce your score for this. Also image optimization DOES matter and has matter for many, many years (will also be included in scores for classic PageSpeed and YSlow too).

The scores of your sample websites are quite accurate and you have naively assumed that similar load times on a desktop/laptop/server with high speed internet and a decent processor will therefore always be fast load times. Take that website with a 2.7mb download and 55 http requests to a 3 year old mobile phone on 3g internet and watch your load time hit 30 seconds or more! Take the other sample website at 418kb and 23 requests on that same 3 year old mobile phone with 3g internet and the load time will be a tiny fraction of the comparison (probably 25-30%). So the page speed scores are very accurate indeed and DO reflect the overall page speed. Remember as of this last April google is focussing more strongly on mobile!

You state that “Google PageSpeed doesn’t even measure the loading time of your site?” Can you please site the credible source where you got this information?
Thanks

Great google speed insight article. I guess I’m guilty of putting too much emphasis on this metric. Good to know I was wrong for future tests.

Thanks.
Jimmy

Thank you for this post. I have been fighting tooth and nail with clients trying to defend the exact same ideas you posted in this article. I even came up with a hypothesis as to why Google Pagespeed is mis-leading to the general public. Recently I wrote an email to a client to try and help provide more insight to the Google Pagespeed tool.

Client,
To compound on the recent email regarding Google’s Pagespeed tool I recently came up with a great way to debunk Google’s Pagespeed tool and it’s misleading scores.
Try it out for yourself. I figured, hey what would Google’s Pagespeed tools show/say about the auto manufacturer websites. I mean it would be a given that these high profile websites must have the best development teams and surely there are no big issues when visiting these websites, right? Well I ran a couple of tests on Ford, Chevy & Dodge. Needless to say the scores would reflect that these sites are completely failing. yet we know they aren’t. Both design usability and page loads are just fine on my computer. (* comcast land line, desktop, using safari)

https://developers.google.com/speed/pagespeed/insights/?url=http%3A%2F%2Fwww.ford.com

https://developers.google.com/speed/pagespeed/insights/?url=http%3A%2F%2Fwww.dodge.com%2Fen%2Fchallenger%2F&tab=mobile

https://developers.google.com/speed/pagespeed/insights/?url=http%3A%2F%2Fwww.chevrolet.com

So why would this be the case (the fact they are going failing page speed scores)?

Heres my take on Google’s Pagespeed tool. It’s a failure right now (in the sense it’s being mis-represented). Some businesses & people are out there using it to create a fear that somehow websites are failing or need major changes to become compliant with Google. That somehow a vast overhaul of what you are doing online needs be changed or else you & your website may fail. The fact of the matter is this tool is for Developers and is a tool for developers to utilize to figure out ways to cut server load and reduce server/hosting expense. The real kicker with this so called tool is that some of the points it will deduct you for go against web standards protocol. And Google is part of the web standards initiative so it is mind boggling to me how they (along with microsoft and many other companies) are suppose to push for this so called web standards to make it easier for designers and the end user but then develop this tool which makes recommendations contrary to the so called web standards.

I’m not convinced this tool was ever meant for the general public or general business to business discussion. It has its value & purpose but I think when people start showing the Google Pagespeed tool around like its suppose to provide some sort of insight to folks who aren’t developers, it certainly makes the waters murky and more confusing.

As professional web developer I can’t agree with most of things written in the article. It is obvius that author of the article never built nothing, or have lack of knowledge on how to properly build a website. While I agree that pagespeed insight does not concentrate around speed only, it does a lot more. Also another thing is I see that author of the article used another “generator” that have very limited approach location, which proves how big amature author is. Ofcourse that if you load a website which server is located in Afganistan and your starting location is Canada ofcourse it will load a lot more. Proves that author of the arzcile is missing some crucial and basic knowledge about networks. Second, for myself I use site analyzer with google insight. Doing things right and by the standards will keep your projects healthy and on top of the class. Others, as the author of the article without basic knoledge on how to properly build a website will whine.

@Ben

If you read this post carefully, you will notice, it not sad to does not care about your health of your webpage. It says, do everything what you can, but after that do not spend more time and money to to try to incrase your page speed to 100m or with 1 point.

This PageSpeed is a bullshit.

Some of my experience, and impressions:

Clients always want everything, fast, good, and cheap. They want to monitor they visitors. What is the best tool for this? Google analytics.

From this point, you are lost, and your site will be never score 100%! It is also true for jQuery.

I am loaded it form googles CDN, and they sad to their own script, to use gzip??? What the heck? It is so ridicuolus.

Javascripts: they always say, to defer / asynch it. It is impossible to. Now I am developing a jquery mobile site. jquery mobile javascript is depends on jquery framework, so you can not load it asynch.

The only way, to earn the 100 points, if you create a simple HTML page with no js and no/very little inline css.

This is also true for AMP pages.

Speed Junky – your http://fastwp.de/ site has a 100/100/97

You should work on prioritizing visual content.

You also aren’t using google analytics.

Props to you, though, for having a wordpress site and receiving almost perfect grades.

check out http://iqrco.de

That is the only grade so far I have seen with a 100/100/100 on Google Page Insights.

On GTMetrix it has a 100/99 because it doesn’t use a CDN.

This is a terrible article, and it spreads some pretty bad mis-information. Why so focused on load times?

They are far far far from the most important metric. The most important thing for your users is how quickly the site is usable; ie, what pagespeed measures. You need to worry about speedIndex, not page load. Why are wordpress devs so bad at this?

You wonder why pages with long load times get higher scores? Who cares if the site takes 10 seconds to load if it’s usable within .5 seconds vs a site that loads in 5 seconds but isn’t usable until 3 seconds in? These aren’t just made up cases – it’s common. And the solutions are easy so why not do them?

TLDR: Perceived performance is the most important thing. Don’t block the critical rendering path. Learn to read a waterfall. Load times don’t matter.

Read these articles instead to get this nonsense out of your head:

http://www.sitepoint.com/speed-index-measuring-page-load-time-different-way/

http://www.stevesouders.com/blog/2012/02/10/the-performance-golden-rule/

http://www.stevesouders.com/blog/2013/05/13/moving-beyond-window-onload/

https://speedcurve.com/blog/critical-blocking-resources/

    @JP: Why you should be focused on the page loading time? It’s the most important metric you should care.

    If you have 100% at PageSpeed, it doesn’t mean your website is fast. It just mean you implement some recommandations to “help” you to speed up your website. It can’t assure you to have a fast one.

    You can have 100% and a page loading time to 5 seconds or you can have 30% with only 500ms to be loaded. You can have 100% on a website, if you are on a shared hosting and you pay it $3 / month, you can be sure that your page loading will be still slow.

@JP

I really hate those sites, what is usable in 0.5 seconds. This is, because site is loading very fast, but not the adverts for example from CDN. And most of the webmasters do not use placeholders. So, when the title of article loaded in 0.5s, I am happy, and I want to click on it. But while I move my finger to the title, and click on it, an ad or a picutre from a previous article is push down my title, and I click on something else accidentally. Then I lost some traffic, I need to push back button, and do the whole procedure again, but this time I am waiting that 5sec to be sure, there are no more element to load. So the “above the fold” is good only, if there are placeholders. But if there are, you will see a big white area.

But unless you are willing for high end google search results you need a high score on pagespeed. Unless I am mistaken.

    @Jesse: Your PageSpeed score doesn’t have any effects on the Google search results. The only metric Google cares is your page loading time.

Thanks to all for sharing their own experiences.

If you read information carefully from Google Page Speed Insights you will find that they don’t care about server time response. All they need is your page speed and the user experience. This is one diference between Google and Pingdom Tools.

The truth about Google PageSpeed is that your score matter and you should do the best to get hight score. I’m Front end Developer and I’m specialised in optimize websites for Google Page Speed. For example my website have 91/100, 97/100 hight score, I improve this score to make both users and Google more happy.

In 2016 you will never find websites with low speed or bad code in the first results. You should care about Google Page Speed.

WP Rocket Support Team

@Dan: Do you really think that Google Bots will take the time to make a Google Page Speed test on all your pages? It will cost them to much money to do that. It’s why the Google Page Speed grade can’t be care for Google. And actually, the page speed doesn’t have any effects on your SEO rank.

Hey Jonathan,

Tx for your reply, we can test this. I will make a page with some poor/medium loading time, and after that I will optimize using Page Speed to get highest score. No other changes. Did you want to guess who’s gonna win?

Also, why wp-rocket.me has all the html, css, javascript compressed if page speed doesn’t have any effects on SEO?

If you can have you website faster, do it! users will love this, and if users will have great experience, Google will be happy.

Cheers!

    WP Rocket Support Team

    @Dan: We aren’t saying the “Page Speed” has not effet, we are saying that “Page Speed grad” has no effet, it’s different. The page loading time is the most important metric, the page grade doesn’t have any effects on your SEO!

thats not very smart, if the pagespeed is bad the bot will stop before he have seen your fuill side as faster the page so more he can index. if the speed is very slow you get penalties because the user expirience is bad…

    WP Rocket Support Team

    @cerianpalma: Google Bot doesn’t make any “Page Speed grade tests” on your website. It crawl your website and measure its loading time. Google Bot doesn’t care to your performance note.

Good article. I definitely improved my website quite a bit back when I was new to this kind of stuff by following some of the PageSpeed guidelines, but like mentioned in the article, some are impossible or even ridiculous 🙂

nice vision of the Google speed strategy. But I consider it matters, as i’m trying my best to move at least little bit our website http://www.pnnsoft.com/ by for the last 3 years it remains at the same point. We improved everything, but still the Google PageSpeed is low, we cant identify why. I think it is the reasoon a website is moving soooo slowly. I thik so, as the same topic website is mow visible by good keywords, and the website is only 1 year old))) abd the Google PageSpeed is good.

Any suggestions?

Thank you for the step by step guide. Through your guide I was able to achieve good score of avg 96% on pingdom and google page speed. My only concern is with the speed site speed. I have tried both cloudflare and cloudfront and none worked. My question is can we use both of them?

We used the Google Pagespeed Insights service to optimize our website performance on mobile devices. It helped greatly, we’ve managed to improve user experience greatly.

Thank you very much, I have wasted many time to gain Google Pagespeed score. Now I’m focusing only on Page load time.

Great post Jonathan. It’s true, most of us use Google Page Speed Insights as a tool. As Lubos said though, the above the fold load time is more important than the load time of the whole page. Why else would Google tell people to inline the above the fold CSS?

The point is to deliver the best user experience. That is is what Google is aiming for, not the score. Page Speed Insights is a guideline for great site performance, not the wirtten-in-stone law that says your site won’t rank if you have a bad score.

There are actually many sites in the first page results with a failing Page Speed score (for now). Test it and see.

Google currently checking the mobile friendly and page speed issue, therefore its really important to improve the user experience. As a web designer, I always give priority on page speed when I create a website. I have few website with page speed score about 90 and I am getting better result.

I usually check how the competition does in Pagespeed insights. I check the first 10 results in Google search.
I rarely find a site with a score over 60 in mobile or 70 in desktop. And the ones that have good score are sites with old design, some are not even mobile friendly (no responsive or mobile site) but they get scores over 60 in the mobile grade.
How on earh it is possible a site that googles finds to be not mobile friendly to have a 70+ score in the pagespeed mobile category is beyond my understanding.
So well.. PageSpeed is just another mean to make people spend their money for nothing. And make the honest developers loosing clients who trust google’s wisdom more than common logic…

Try the free website speed tests https://loadfocus.com/website-speed-testing, the best page load test tool I’ve found so far.

WP-rocket absolutely rocks. My website (www.heirloom.photography) is going 4 times faster !!! Just one question though, how do you know which js files to put at the footer?

    WP Rocket Support Team

    @Alex: If you turn on the JS minification, WP Rocket will automatically puts at the bottom all JS files declared with wp_enqueue_script to be in the footer

Yeah false.. my wordpress site is extremely graphical and jquery reliant and i have 100/100 mobile and desktop. Its not impossible you just need to know client side and server side programming. To many people trying to say they are web designers when they use templates or throw together some spaghetti code they found googling the how to tutorials. For a site like this to say its nearly impossible is scary.

I have done my own testing over 6 months. My site is booming in the serps on mobile and desktop since changing to 100/100 score and that is the only thing i change. Looking at the analytics i can see the change start to climb as google reindexed my changes on each page.

It hit its peak and i went back and got wc3 compliance 100% too and again got a large spike in serps and more traffic again.

Thanks for this post. I find that google makes a hoot and holler mainly about page load speed. I will take note of what you speak about here in this post however and analyze it to see what I can work on for my own website. Thanks for the post.

Your article is misleading. You say Page Speed score does not matter… what you need to make clear is page speed score does not matter for actual speed of the site. It does matter for other aspects, including the fact Google takes it into consideration for ranking. Your article implies it just does not matter at all. So you should make it clear that does not matter for page speed.

Although of course it does, because optimising images, CSS, HTML, increasing cache time, and many other factors, actually WILL result in a faster site 🙂

Hi Laurence, I am not sure the google bots spiders score our sites using Google Page Speed. In my opinion, the do NOT consider Google Page Speed score, but the REAL speed when they scan the site.

Hi Lucy, thanks for clearing out myths related to Google Page Speed Insights. You’re right, it’s not the end if we didn’t got 100/100 as some issues may not be solved or not in our control e.g. third party stuffs. So, it’s best to solve only those issues which is in our control and those who don’t affect our website in anyway.

Today I managed to score 98/100 with 100% user experience in both mobile and desktop views on my website http://www.earnext.com

Let’s see if it makes a real difference.

I very insightful post on Goggle Speed Test Insights! Very well written and actually helpful.

You may use a service like this
https://jonassebastianohlsson.com/criticalpathcssgenerator/
for finding the render blocking CSS to extract the above the fold css parts, then use them between the head tags. If someone is interesting to see the result check my website, I have implemented it with autoptimize. It gives a better result, and a faster speed.
http://cuta.info/

Hi Lucy, thanks for clearing out myths related to Google Page Speed Insights. You’re right, it’s not the end if we didn’t got 100/100 as some issues may not be solved or not in our control e.g. third party stuffs. So, it’s best to solve only those issues which is in our control and those who don’t affect our website in anyway.

Forget about this page speed. I’ve seen sites that take ages to load but they rank at the top of Google page 1.

Thank you very helpful article, I’ve been following of the Google PageSpeed guidelines.
I got 100/100 mobile and 100/100 desktop

https://developers.google.com/speed/pagespeed/insights/?url=https://ideatoz.com/

There is an addon plugin for WP Rocket on the wordpress site that will remove render blocking css by adding Async: “WP Rocket ASYNC CSS”

And also remove render blocking JS: “WP-Rocket Footer JS”

Ok so we can already add JS to footer manually with WP Rocket, but why no option for Async or ability to insert critical css? We have to use other plugins for this.

The really sad thing is that page speed test seems to have absolutely no bearing on organic search results ! ! Just ran a test of the top 3 results (after the ads) on several searches and none of them came close to passing the test ?? Since I am in Ecommerce I am constantly searching for products on Google, but their results are getting worse every week and no matter what they claim, seem to be very much influenced by advertising budgets… Do I really need to be told that Amazon, Ebay, Target, Jet, aliexpress and Walmart have frying pans for sale ? I already knew that ! If i am searching Google I am looking for the alternatives to the obvious !

“Perfect scores aren’t possible”.
I accomplished it with blogger. It took removing a lot of the native code and using lightweight javascript to accomplish the same effect as heavyweight JQuery and such. Getting a Pagespeed 100% for awhile now, and average reached of 95% to 98% on YSlow on GTMetrix reports. Even reached 100% a time or two for YSlow. Speed ranges from: 0.5 to a couple seconds, max.

This article is wrong

It’s easy to have 100/100 on google, and 99/100 on most of website speed tests, or 1 grade !

Just learn how to build a website correctly, and you will have those stats !

When i do a website for a customer, i don’t loose my time by trying to get those stats, those stats are natural when the site is building in a good way !

I’m a Computer Graphics researcher and my site strongly depends on images. In my case, the only way to make Google PageSpeed Insight happy about the images is using the “optimized images” it generates itself. The problem is: Googles’s dummy image optimizer degrades image quality in many cases and I have many examples to prove it. I will not discuss the tech details in a post comment, but It’s so serious that I think it deserves an exclusive article.

This is very very helpful post for me, i was very confuse between google page speed and Gt Matrix tool. Both are showing huge difference in page speed. But now i can belive only on Gt Metrix which showing B grade of speed. Thanks for clearing the doubts.

I reached 91 mobile/95 desktop on google’s pagespeed test, and 100/91 on gtmetrix using theme.co’s X-theme, wp-rocket and autoptimize. I’m not a developer, and my site isn’t fancy, but it’s not stripped down, uses jquery and google analytics.js.

https://schecteracademicservices.com/

So it’s doable. Page load times .5-.7 sec.

Wp-rocket is awesome, definitely worth the $, but I do WISH that I could defer/async css with wp-rocket, because then I wouldn’t need autoptimize to handle that function.

Currently Google PageSpeed Insights is unusable. Shows wrong results.
Launched on February 2017 the tool it buggy now.
The Google team said they are working on it…
Watch pagespeed insights forum.

A Google PageSpeed 100 score has a significant effect on SEO. I’ve seen many customers, sometimes with a new website, obtain top 10 rankings for premium competitive keywords and often receive free business box promotion on Google mobile (#1 spot in local business, the physical #1 spot on modern Google pages + a large right side box with company details, pictures and more). The reason for this is that Google is increasingly offering users dedicated mobile services, for example to find a local business, and when Google has to select a website for use in their services they will select a website that has the best chance to offer users an optimal (mobile) experience, which is measured by the Google PageSpeed score.

The Google PageSpeed score is a grade that shows how Google perceives the overall reliability, speed and user experience quality of a website. Google’s intend is to make money by satisfying searching users. In order to become the primary result in the modern Google, it basically comes down to making Google believe that your website has a higher probability to satisfy the most users for a search term. A Google PageSpeed 100 score is a factor that tells Google that the technical / user experience quality of a website will be optimal so that the chance will be high that users will be satisfied with the result in diverse (complex) conditions that could include mobile device, environment, internet connection speed (history), geographic location and more.

It is true that pagespeed is important.
But for me Google Pagespeed Insight score provides an analysis on how the performance speed of my site. I do not need to get a perfect score, the more important is the real speed of my site satisfying the visitors.
After all, many of the top sites have Google Pagespeed Insight score only 85/100 or less.
Facebook = 68/100 and 88/100
Amazon = 65/100
wikipedia = 86/100 and 94/100

Fun fact 1:

Create a fresh page with a google AMP validated source including just a question mark in the body – yes thats right – pretty muich the boilerplate and and 1 byte for the document. No images, no css no javascript (not allowed anyway).

*drumrole*

Score: 91

That did it for me, screw google pagespeed!

Fun fact 2:

Adding lots of code and several images and javascript files I managed to get a score of 96 which goes against logic given a ? boilerplate gives me constant 91. I have experienced pagerank for being unstable earlier but know it is – how they magically calculate their number is obviously due to some luck in how you stack your elements even if you add to the page which again – as a programmer – I can not have anything to do with such a service as it melts my brain in agony!

Great article. The speed of the site is very important thanks to topic .

Hello from 2017! You can ignore the Law (read Google PageSpeed score) and live by your Own Rules, based on personal conclusions. But only at home. Otherwise you’ll get punished. You can also argue that the Earth is square, the sugar is salty and the dogs are able to fly. Who cares?

But if Page Speed Insights is telling you that there’s a huge delay in your server response, css and js blocks visible content appearance etc… Imagine your giant page loads in 4.3 sec and (!) first visible data appears only after 3 sec.. users will go elsewhere. It’s all about UX nowadays

Nice article, clear up. thanks wp-rockets.

Yes. Even I dont think page speed will be going to have impact on ranking. Google page itself has less page speed LOL!

The speed of the site is very important thanks to topic Great article.

I am slightly skeptical about properly using the Google page speed insight tool as it display the lines of codes to minify the CSS and JS codes to improve the performance of the website. With the use of this Google tool came to know how to improve the page speed of the website. On implementing the same for website http://www.infirays.com has helped to improve the overall loading speed and ultimately rankings. I would say everyone has their own way of experimenting the things with big G.

I agree with everything in this article except the title but only in the case that the big G actually uses the same logic as the GPS tools to grade websites it indexes.

If this is the case, we should actually all BE REALLY CONCERNED because our sites are often being graded unfairly because the tool lacks the proper depth to actually know and understand a real-world human experience. I get upset of the idea that my sites may rank poorly because a tool used to grade it isn’t up to the task logically. Funny thing is when the GPS tool is used on the big G’s own sites, it doesn’t get good grades either half the time. Again, grain of salt yes but if we’re being unfairly graded and this impacts our ranking then we should actually be pretty upset about these tools and demand better.

I have sites in which the full visible page load is in low ms but still get outrageous scores in the 60-70 range while using other tools I get extremely high grades. Often the suggestions made by GPS have nothing to do with the actual experience a real human would have.

I lobby to demand that Google fixes the logic to understand how websites in 2017 are built, not just against the most common methods used a few years ago. We have websites in which the DOM is entirely full-duplex javascript applications. Also, the image compression often suggested are of terrible quality yet it complains about a few bytes with big mark down results when it comes to overall score simply because you did not use the “optimized” images they provided. The fact that a few bytes change a score by such large numbers tells me they don’t have their logic right.

For example, techniques such as lazy loading or content hidden behind modal triggers often count against you because the tools cannot perceive what the human actually sees vs what is left behind the scenes as part of the user interaction later in the experience.

I’m glad you concur. I have come to the same realization about ‘chasing a grade!’ The image compression seems to most always be the leading factor of page speed. This, thank god is probably one of the easiest DIY corrections anyone can make! Thanks for the article!
https://www.finneganseo.com/harrisburg-seo

Hi,

This is an article give me some oxygen, since last week I am in huge confusion and trouble after disapproved my google ad sense application . I try to optimize my site as per google insight speed result which was showing 90 for desktop and 74 for mobile. After long try and activate speed booster pack plugin it improves 96 for desktop and 95 for mobile.

But now another issue raised with the appearance of the sites. Can you please suggest whether I revert back to my old position and set up, is it really matter to get the google absence approval? Or any solution further to optimize the appearance issue?

I always thought that the speed of the PageSpeed is the speed of loading the page, thank you, you opened my eyes !!!

Please I have the plugin installed but I keep getting everything wrong on google tests. My pingdom score is B. I don’t know if thats Nice. Please can you test and give my website recommendations on what to do. Here is the link http://www.cybergeak.com . Thanks I really expect a helpful reply from you soon…

I didn’t trust google pagespeed test, it’s not accurate. If we tried to test “deviantart.com”, the score is 1/100, OMG this is the worst score I have ever seen. But if we open the actual site it took 1s to load.

On the other hand, the score of my blog is 50/100, not good but it’s better compare to deviantart score, but in real live my blog took 6s to load, which is very slow.

Btw my website is very small only 2000-5000 UV/month, hosted at Linode 4GB plan, I’m using keyCDN and cloudflare. Last year fully load only took 600ms-1s, then I installed SSL and social locker, I think this plugin is very heavy, my website become very slow.

Currently I’m using W3TC, but I’ll happy to try another plugin, do you think Wp Rocket plugin can help my page speed? I really missed the fast speed like before.

Thanks 🙂

    WP Rocket Support Team

    @Kris: The best way to know if WP Rocket will speed up your website is to try it. We have a 30 days refund policy so you can try it without any risks 🙂

Thanks for this awesome article.

Now I understand that Google PageSpeed Insights has no importance for Load time or SEO.

So I am using only WP-Rocket on blog https://www.earngurus.com/ and checking my speed on GTMetrix and Pingdom

Nice article! I wish I could have read this article a bit earlier because significant time has been wasted on Google PageSpeed Insights.

It does indicate some potential issues with the site but really not worth the effort to hit a certain score.

Oh mei, thank god a friend of mine has sent me this article! I was quite a while going nearly nuts with all this speed stuff. Working in all corners on my website, http://www.anettemossbacher.com, to get a Ferrari running, if not a Formula One racing car. Now when I read this, that subject matter is going to be closed. We have 2 websites, my own and an online shop. The online shop runs with WP-Rocket already, it will be more fine tuned. A Cache plugin is needed indeed for websites. For the Leverage browser Cache – Google analytics – there exist a plugin to get rid of it. Honestly I did not see much improvement when using it, it only disappears in the page speed test results! Never mind for such, I my opinion. To make images load fast in a website is a photographers challenge for sure. There are some great tools, yes Photoshop…etc. imigfy for in the WP site… There is a need to get images load fast, with the sacrifice losing on image quality. Bad for a photographer! However, I was so glad to read this article, thanks so much. I told already my friend who sent it to me my Million thanks.
Have a fantastic day and for sure I am now cured of the page speed syndrom 🙂

Ciao Anette

Nice article but have you check the crawling data after such an upgrade of your score? We have seen a significant increase of crawling on our website may be due also to the change of server too. I will survey the data for a couple of weeks and report the results here. NB we have removed the “Eliminate render-blocking JavaScript and CSS in above-the-fold content” at the moment so the score drop to 87/100. And for an unknown reason, we have a different score for mobile if someone has an explanation for this? Check the website on pagespeed http://dlcompare.fr and give me your toughts.

That was real help. I was chasing the speed grade like mad.. Thanks I found your article.

This is a very useful post. I do use pingdom tool, gt matrix and google page insights to analyze my blogs

thank you for the tips and heads up

Great Article. I found this whilst looking for Page Speed Insights, don’t think I will bother going there now lol

Focus on the actual site load time of a site, since Google PageSpeed is just based on the rules that are set.

https://developers.google.com/speed/docs/insights/rules

Great article, I’m using Google fonts on my site and insights is coming back saying that the css needs to be optimised! You’d think they’d make sure their own stuff was done properly if they are going to penalise websites that use it.

you are not right. I got for https://beerm-ksa.com 100/100 for first page and optimized most landholdings . This is new website and it is ranking fast . If you get 100 your page will be faster just because less http requests and smaller code.

That was real help. I was chasing the speed grade like mad.. Thanks I found your article.

Great article, I’m using Google fonts on my site and insights is coming back saying that the css needs to be optimised! You’d think they’d make sure their own stuff was done properly if they are going to penalise websites that use it.

The website speed is really matter that’s why we had optimise our website. On PageSpeed Insights we have 100% score on both mobile and desktop. And yes it is difficult to do it but not impossible.

Leave a Reply

Your email address will not be published. Required fields are indicated by *

[i]
[i]