Test tools for Websites - post them here!

[rth_discos]

In terms of speeding up website loading times, I've found this plugin to be the best as speeding up the site (and I've tested a fair few)

https://en-gb.wordpress.org/plugins/wp-fastest-cache/

It's always made an immediate difference

Essentially, turn on every setting that's available in the free version, except for "Mobile: Don't show the cached version for desktop to mobile devices"

[Marc J]

The main contender on all my research though was the host - it took 9.5 seconds for your one to respond, the second test went to 13%! This is the first thing any agency will say and it will cost you hundreds to put right, especially if you start moving hosts and they manage it for you. You can save some cash by doing it yourself, but for ease/professional service maybe an agent is the way forward. :)

Just picking up on this comment as while you've pointed the finger at the host (myself), this isn't the issue, i.e. simply moving hosts won't improve the score Google is giving it. Sites on the same server give server response times in the same test in under Google's recommended 0.2s. The "server response time" in that test is an umbrella for a lot of things - basically everything needed before the page is able to render to the browser.

This includes any and all HTML / JS / CSS required, which may come from external sources (in this one, mostly Google Fonts which are normally OK, but Instagram and tawk.to are adding almost 1.5 seconds).

The more complicated (or unnecessarily bloated, or poorly coded) a site is the longer it'll take to load. There are cache / minify / compression / CDN settings between the server setup / website code / Wordpress plugins that'll improve things, but that general rule remains. I currently recommend W3TC (https://en-gb.wordpress.org/plugins/w3-total-cache/) and Cloudflare as a good starting point for Wordpress sites (see http://www.bloggingwizard.com/w3-total-cache-cloudflare/).

To get Wordpress to run super fast, you'd normally be looking at your own VPS server so you could set it just as you need it, running Litespeed instead of Apache (https://www.litespeedtech.com/benchmarks/wordpress) using HTTP2 (https://www.litespeedtech.com/products/litespeed-web-server/features/http-2-support) and then tweaking further, so you'd be in to around the £150 per month mark (but you could host many sites on that). You'd still need to make sure your WP sites were as quick as they could be, though - no bloat!

Just one more quick comment re http://www.spiritshigh.co.uk/, the login form (accessed through the navigation on every page) causes Chrome to say every page is not secure as there's a password field on the page. Either add an SSL (which would mean losing Cloudflare as it's not compatible) or change the login nav item to just a link, so that only the login page has the password field on it.

[Marc J]

Just to add to this, I had a closer look and the Instagram images are all for the widget in the footer, which while showing the 12 most recent images from Instagram at 50x50, loads them at 640x640 and then resizes them with CSS.

The 12 large images add 874Kb to the page payload, whereas if they were loaded at the size displayed they'd be a total of 20Kb for all 12.

It's as simple as changing the URL for the image, from (for example) http://scontent.cdninstagram.com/t51.2885-15/s640x640/sh0.08/e35/18512714_1210432885733649_6686078849286930432_n.jp g to http://scontent.cdninstagram.com/t51.2885-15/s50x50/sh0.08/e35/18512714_1210432885733649_6686078849286930432_n.jp g

Note the change from s640x640 to s50x50 in the image address.

I don't know where that's set, maybe in an Instagram plugin you have installed, or in the theme widget somewhere, but if that can be changed it's a simple change that would make a big difference.

You should also optimise the images that Google is saying should be done (the ones hosted locally). You can download them already optimised from the pagespeed results page (just let me know if you want me to do it for you, FOC ;) ).

[Marc J]

OK, I tweaked the Instagram widget code (wp-content/plugins/wp-instagram-widget/wp-instagram-widget.php) to size thumbnails at 50 x 50 rather than 160 x 160, and changed the "Photo Size" in WP Admin -> Apperance -> Widgets -> Footer Sidebar #2 from "Large" (which was the 640x640) to "Thumbnail" (which is now 50x50).

I also uploaded all the optimised images Google was recommending (with the ones they provided), so the whole images part of their recommendations should now be gone (for the Home Page, at least) ;)

Hope this is OK, the original images are still there appended with ".original" if you need them....

[Spirits High]

Just to add to this, I had a closer look and the Instagram images are all for the widget in the footer, which while showing the 12 most recent images from Instagram at 50x50, loads them at 640x640 and then resizes them with CSS.

The 12 large images add 874Kb to the page payload, whereas if they were loaded at the size displayed they'd be a total of 20Kb for all 12.

It's as simple as changing the URL for the image, from (for example) http://scontent.cdninstagram.com/t51.2885-15/s640x640/sh0.08/e35/18512714_1210432885733649_6686078849286930432_n.jp g to http://scontent.cdninstagram.com/t51.2885-15/s50x50/sh0.08/e35/18512714_1210432885733649_6686078849286930432_n.jp g

Note the change from s640x640 to s50x50 in the image address.

I don't know where that's set, maybe in an Instagram plugin you have installed, or in the theme widget somewhere, but if that can be changed it's a simple change that would make a big difference.

You should also optimise the images that Google is saying should be done (the ones hosted locally). You can download them already optimised from the pagespeed results page (just let me know if you want me to do it for you, FOC ;) ).


OK, I tweaked the Instagram widget code (wp-content/plugins/wp-instagram-widget/wp-instagram-widget.php) to size thumbnails at 50 x 50 rather than 160 x 160, and changed the "Photo Size" in WP Admin -> Apperance -> Widgets -> Footer Sidebar #2 from "Large" (which was the 640x640) to "Thumbnail" (which is now 50x50).

I also uploaded all the optimised images Google was recommending (with the ones they provided), so the whole images part of their recommendations should now be gone (for the Home Page, at least) ;)

Hope this is OK, the original images are still there appended with ".original" if you need them....

Marc,

You are a Superstar! Thanks for doing that feel free to do anymore bits if you want :D

I've run it through "Toby's Checker" and it now scores 46 Mobile (was 19) & 59 Desktop (was 25) Incidentally, now better than yourdj.co.uk!!

This is why this forum should never be let to get as close to disappearing as it was recently . The font of knowledge on here is outstanding and I could have no doubt been charged £000's for a few little tweeks that our lovely members have offered up for free!

[Shaun]

Nice one Marc. :) :beer1::beer1::beer1:

[yourdj]

I've run a few sites through that page insight checker and even the "big boys" like Argos and Wiggle don't fair aswell as I would have expected so it's something worth looking at.

Yes I guess, although people know what they are going too, Ebay takes forever sometimes, but I still wait for it ;)


In terms of speeding up website loading times, I've found this plugin to be the best as speeding up the site (and I've tested a fair few)

Thats the best reported free one. I use this.


Just picking up on this comment as while you've pointed the finger at the host (myself),

Oh sorry Marc, assumed it was a random cheap one. :)
Well done for investigating it, i am no expert, just judging from my own experiences this week.
Out must be nice to have an actual person helping out, rather than a techyy in a call centre.

I did have a tech guy, but he has moved on to bigger and better things now.



I've run it through "Toby's Checker"

Its not my test, its googles website analytical tool. Probably the best free one i have come across. :)


46 Mobile (was 19) & 59 Desktop (was 25) Incidentally, now better than yourdj.co.uk!!

I have had the caching software off today/yesterday. it wont allow me to edit the site with it on for some reason. :)
Its gone down a bit to 67 & 70 but I have added a few bits to the footer, so thats probably why.

Thats a much better score, if you have enquiries monitored, then the results in a month or two would be interesting. :)
My guess is people were just getting bored and clicking off before it loaded. Its a great site so you should get business from it.

[Marc J]

Oh sorry Marc, assumed it was a random cheap one. :)

:eek: :daft: :muppet:



I have had the caching software off today/yesterday. it wont allow me to edit the site with it on for some reason. :)
Its gone down a bit to 67 & 70 but I have added a few bits to the footer, so thats probably why.

Can't have you beating us, Toby....so I made a few more tweaks :p

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

18348

[yourdj]

:eek: :daft: :muppet:



Can't have you beating us, Toby....so I made a few more tweaks :p

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

18348

Nice. :)

I just added that fastest wp thing to my HDA site and look what happened!!
I also updated all the plugins, so its not necessarily the WO fastest cache.

From this:

18350

to this:

18349

Looks like it may be a conflicting between the theme and the plugin??

[rth_discos]

Whilst we're testing website speeds - Google gives a "ranking" - but doesn't indicate load speed.

Worth also running through this (where you'll see the biggest results that a cache plugin achieves)

https://tools.pingdom.com/

Use Sweden as the "test from" location as it's most local.

I'll start the benchmark with my result:
18351

[Marc J]

I'll start the benchmark with my result:

Is that a gauntlet I see? ;)

18353

Paul - I don't know how much use your tawk.to feature gets, but it's causing issues on both tests and it might be worth ditching it if it isn't used. I looked all over the page and couldn't actually find it (I'm assuming it's some sort of live chat).

Another good test is https://gtmetrix.com, but you need to create a free account if you want to test from different locations (i.e. London), as the default is Vancouver.

[Imagine]

Go on then - I'll rise to the challenge :)

18354

My biggest negative on all of these sites is that I apparently need to leverage browser caching. Interesting because I went and purchased the pro version of WPCache :o

[Spirits High]

Paul - I don't know how much use your tawk.to feature gets, but it's causing issues on both tests and it might be worth ditching it if it isn't used. I looked all over the page and couldn't actually find it (I'm assuming it's some sort of live chat).


It's not on all the time and is mainly switched on the day after a Wedding fayre especially the big fayres. Ideally it'll be on all the time but I can't always guarantee I'll be sat at my pc. I'm in the back end now so will go shut it down!

[Spirits High]

Following on from my Media agencies thread which has now been well & truly hi-jacked as a test bed and web ego trip thread I've moved them all over here.

If you have a web speed / page insight type tool you use post it here along with your result and we'll see who rules supreme!

[Marc J]

My biggest negative on all of these sites is that I apparently need to leverage browser caching. Interesting because I went and purchased the pro version of WPCache :o

Nothing you do will fix that, looking at the Google PageSpeed Insights results for your site, the "Leverage Browser Caching" relates to the following URLs: -

http://www.djep2.com/clients/includes/stylesheet_updates.css (expiration not specified)
https://static.doubleclick.net/instream/ad_status.js (15 minutes)
https://connect.facebook.net/en_US/fbevents.js (20 minutes)
https://connect.facebook.net/signals/config/193444594352069?v=2.7.10 (20 minutes)
https://maps.googleapis.com/maps/api/js?v=3.exp&libraries=places&key=AIzaSyAaWMXkbWfqulqZ-9OzwwZSgKP2nQVftfA&ver=4.7.5 (30 minutes)
http://i.ytimg.com/vi/0UOoLlparmc/hqdefault.jpg (2 hours)
http://i.ytimg.com/vi/JBGZpzQHVV8/maxresdefault.jpg (2 hours)
http://i.ytimg.com/vi/S1ZdcT0hBuI/maxresdefault.jpg (2 hours)
http://i.ytimg.com/vi/XJa0ZiF9_Nw/maxresdefault.jpg (2 hours)
http://i.ytimg.com/vi/X_1KPsHCEzU/maxresdefault.jpg (2 hours)
http://i.ytimg.com/vi/gyXmXGK0Juc/maxresdefault.jpg (2 hours)
https://www.google-analytics.com/analytics.js (2 hours)

None of which are on your server, so they are all outwith your control.

[Marc J]

It's not on all the time and is mainly switched on the day after a Wedding fayre especially the big fayres. Ideally it'll be on all the time but I can't always guarantee I'll be sat at my pc. I'm in the back end now so will go shut it down!

Much better :)

18356

IMHO you should keep it off, and only switch it on when you're there to deal with it if it gets used.

[Spirits High]

Much better :)

IMHO you should keep it off, and only switch it on when you're there to deal with it if it gets used.

Then that's what I shall do!

"Spoken Web Yoda has"

[Marc J]

For those who want to "cheat" page speed results because the only results in "Leverage Browser Caching" are Google's own Google Analytics javascript files, you can wrap the GA javascript in an "if this isn't a pagespeed test" check, so for example: -

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');

ga('create', 'UA-XXXXXXXX-1', 'auto');
ga('send', 'pageview');

becomes: -

if(navigator.userAgent.indexOf("Speed Insights") == -1) {
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');

ga('create', 'UA-XXXXXXXX-1', 'auto');
ga('send', 'pageview');
}

NOTE: Don't copy & paste, you need your GA ID in there instead of UA-XXXXXXXX-1 for a start!

Pointless as real visitors still get the GA javascript served without browser caching, so the page speed results are technically correct, you haven't fixed anything!

[yourdj]

I am not doing so good on that one. i would be happy with a high C tbh.
I have a lot of plugins and probably a messy code so its never going to be in the A category.

is this thing accurate as my DJ alliance one now shows 64, but on Google its still 10!!
I am more inclined to trust Google as after all its the one and only company I am trying to get rankings on.

Plus their server is not 2,500 km away from mine. :p

18357

18358

[rth_discos]

Performance grade is a rather arbitrary figure.

The useful number here is Load Time.

You want this under 5 seconds.

Anything less than 2 seconds is very good.

Test it a few times - you'll soon notice your server doesn't respond at the same speed every time - so look at the average result over a number of tests.

Beyond 5 seconds and your visitors start to get impatient - which really affects things when it comes to clicking on more pages through the website.

Obviously Google.co.uk will load the quickest as it only has a small bit of text and a search box. As you load up more images and "plugins" to a site, so it slows it down. There's a balance to be made though. Don't turn off a plugin if it speeds the site up, only to reduce your conversion rate if it plays a key part in selling your service. So check everything on your site - do you need it? How important is it? If it's not important, consider removing it and helping your site to speed up.

[Marc J]

I am not doing so good on that one. i would be happy with a high C tbh.
I have a lot of plugins and probably a messy code so its never going to be in the A category.

I don't know...your reported page size is quite a bit smaller than Paul's (1.1Mb vs 2.4Mb) and your number of requests are similar (98 v 89), so your sites are probably more alike than you think in terms of plugins and code. You should be able to score higher than that. Have you tried W3TC instead of WP Fastest Cache? You could also try a CDN (e.g. CloudFlare) which would probably improve things slightly.


is this thing accurate as my DJ alliance one now shows 64, but on Google its still 10!!
I am more inclined to trust Google as after all its the one and only company I am trying to get rankings on.

I'd go with the Google result and try to get that better.

[Marc J]

An alternative to the Google Analytics "cheat" in post #18: -

https://wordpress.org/plugins/host-analyticsjs-local/

The plugin makes a copy of the GA JavaScript on your server. Google doesn't recommend this (https://support.google.com/analytics/answer/1032389?hl=en), but this plugin also checks daily if there's a new version at Google, and overwrites the local copy with the new one if that's the case.

Seems to work well, it got rid of that category and gained an extra point in Page Speed Insights :)

[yourdj]

Thats a bit better ;)

18360

[Marc J]

Thats a bit better ;)

What was the issue / fix?

[DJ Jules]

Whenever I do look at the webmaster tools for page load speeds I find it really frustrating. The issues on my site that cause Google to mark it down are all caused by:

1. Twitter (stupidly long URL's in their code snippets and warnings about enabling compression)
2. Needadisco verification images (could be compressed)
3. Facebook (warnings about enabling compression)
4. Google Analytics (ironic!)

The additional thing I could do is minify the html... might have to look into that.

Anyone played with AMP? (http://searchengineland.com/get-started-accelerated-mobile-pages-amp-240688)

Julian

[rth_discos]

You know the 80/20 rule...

Be careful not to invest your time in the wrong areas here guys.

These tools are useful to highlight any big issues on your websites. And yes, page loading time does affect user experience, and therefore conversion rates (which can also have an effect on your Google search ranking).

Some of the big issues when fixed can help. But trying to shave an additional .3 seconds off loading time and getting a 100/100 result from Google likely won't make any difference to your sales.

That time could have been better spent writing a blog post or two for your website (which really can help improve your Google search ranking) or re-wording your home page to make your service more enticing for visitors (which can make a dramatic difference to your conversion rate).

Just wanted to flag it up before you start heading down a rabbit hole!