Page load speeds are getting more and more important. Not only can a slow loading site annoy your readers, but it can impact your Google search rankings too. But how do you do a website speed test? Read on to find out about the 3 main players in the site speed testing world.
Where do I go to do a website speed test?
As I mentioned before, there are 3 main tools when it comes to testing website loading speeds. They are Google Pagespeed Insights, Pingdom and GTMetrix. They’re all quite different and give you different levels of detail. I’m going to look at each one in detail below, talking about the pros and cons and general usability.
Google Pagespeed Insights
As most of us bloggers want to be found on Google, it makes sense to use Google Pagespeed Insights to check your blog speed. The nice thing with this website speed test, is that you’re given two scores, one for desktop computers and and for mobile devices.
It’s easy to forget that even though most of us use responsive themes (themes that adjust depending on screen resolution) that they’re not usually identical across all screen sizes. This means that they might not load the different parts of the screen in the same way across all devices and so your site might load more quickly on a desktop screen than a iPhone or vice versa.
In my 5 day free Need for Speed blog challenge, this is the page speed analyser that I show my students how to use. It’s straightforward to use and the problems that it finds are layed out in a nice clean way that isn’t overwhelming to users. In my view it’s a great place to start looking at your website speed.
However, as I’ve used Google Pagespeed insights more and more to do website speed tests, the more I’ve found the results to be quite inconsistent.
For example one of my students taking part in my free challenge found that when she retested her page loading speeds at the end of the challenge, her scores were worse than before she started.
But then I tested her site myself minutes later using page speed insights and the scores were looking really good. When I find that it’s playing up, my next stop is GTMetrix.
When I’m doing speed work on a client’s website GTMetrix is my go to website speed test. For me it’s the most reliable and consistent of all 3 of the page speed analysers mentioned in here.
You get 2 scores using GTMetrix, PageSpeed and YSlow. They both check your site using different rules, some rules are very similar, but they won’t be hugely different scores.
My favourite feature of this website speed test, and the one I make most use of, is the Waterfall tab which shows you exactly what is being loaded and when, and how long it’s taking. I’m going to get a bit techy here (if you don’t want to read my jargon, think about puppies and rainbows until I say when).
**TECH-Y STUFF STARTS**
There are different phases of the resources on your page being loaded. For example, your Google Analytics tracking id. For Google Analytics to work it still needs to load some code from Google.
So your site makes a request and then connects to GA. It then sends the info GA needs to send the right stuff back, waits for the response, and then receives it. This kind of info can be really useful in trying to pinpoint where the speed problem is.
**TECH-Y STUFF ENDS**
Pingdom lays out website speed test results in a really user friendly way. It uses letters and colours to give an overall performance grade. But unless you look at them closely the way the results can be a bit misleading. For example, it can give an overall result of B, but then say that the site is only faster than 6% of all sites tested.
Other stats that form part of your results are the load time which is how long it takes for your site to load. It also tells your your page size, which is the number of megabytes needed to download your page and the number of requests it took to load your page.
As part of the results it also tells you what percentage of sites your site loads faster than to give you an idea of it’s speed compared to other sites. A really odd stat that it gives you, as part of the results, is where the speed test was run from. It’s odd because you actually choose where you want the test to be run from before it’s run, but there we go – it’s there incase you forget. Or decide it’s too much hassle to scroll up.
Website speed test – verdict
My preference is to use GTmetrix because it gives me much more detail on where a site is slow however, I think each website speed test has something to offer.
I think GTmetrix is for people who aren’t put off by some of the more technical language it uses. But to be honest, when it comes to testing blog speed it’s impossible not to use some technical terms.
Even though page speed insights can be temperamental when it comes to giving you scores, it’s recommendations are usually bang on.
Pingdom, whilst being a bit odd with its overall speed score, also gives individual scores for the criteria it checks for. In this instance I would say it’s better than page speed insights.
Google’s website speed test just highlights areas for improvement and doesn’t tell you how badly your site performs against them.