Speed Metrics & Timings

Speed metrics and timing

Page loading speed is a ranking factor for Google, given that 40% of users will leave a website if it takes more than 3 seconds to load, and most users will get frustrated if they’re trying to shop from your website and it’s slow enough to make them relive memories of dial-up internet. 

So obviously, you want your website to be fast, both for SEO purposes and customer satisfaction. However, there are multiple factors that impact the loading speed of your website. That’s why speed metrics tools like SideSpeedBot give you specific measurements of the different timings and metrics that affect your page loading speed.

When you first come across all the metrics it can get a bit confusing. In this article, we’ll give you a clarifying overview of what each measurement that SiteSpeedBot uses means for your website. We’ll also provide you with the ideal values for each metric that you should strive toward.

Page Size

Page size, or page weight, measures the total size (or weight) of your landing page, generally in terms of MB. The heavier the page, the longer it will take to load. In fact, you’ll find that when dealing with speed metrics, less is always better! So when it comes to page size, ideally, your page should weigh no more than 2-3 MB. Anything above that will make page loading frustrating for people with a slower internet connection.

Load Time

Load time is measured in terms of Document Complete or Window Loaded/Onload event timing. Again, less is more. A fast site would take between 1.5-2 second to load, and a very fast site could load in less than 1.5 seconds. As we’ve already mentioned, page loading time that’s over 3 seconds can cost you visitors and customers.

This metric is more important to consider than Fully Loaded Time for websites that contain lots of third-party code, such as marketing pixels, affiliate code, and ads. If you get a result of zero for Load Time, check your Javascript files – they’re usually the culprit in such cases.

Number of Requests

As the name suggests, number of requests considers the number of requests on your page in terms of files loaded. Keep in mind that this has nothing to do with site visitors making requests to your site’s server.

Again, a smaller number is better here as well. If less files are loaded when a visitor first opens your landing page, then loading speed will be faster too. The result for this metric will be between 50 and 100 for fast pages, and under 50 for lighting fast pages. If this metric exceeds 200 for your page, you should work on reducing the number of files loaded when a visitor accesses it. There are a couple of site optimization techniques you can try, including deferring Javascript.

TTFB (Time to First Byte)

Time to first byte measures how fast the page’s web server starts sending data to the end user’s browser. A consistently lightning fast response is 0.1 or below. Good results here are between 0.1 and 0.2 within the country of the site’s hosting server, and 0.2-0.5 internationally (abroad). If your TTFB is higher than that, it’s likely that there’s an existing problem you should address.

FCP Timing (Render Start)

First Contentful Paint (FCP) is the point of rendering when a site visitor perceives the page as beginning to load, i.e. when the page begins to show content in the visitor’s browser. FCP basically measures the time it takes for a user to be able to view the content of your page, be it image, text, video.

You want your FCP value to be under 1.5 seconds – 1.5 seconds is what’s perceived as fast by users, and it’s also the time limit that Google uses to label a website as fast. In fact, the FCP for at least 75% of your page visitors should be under 1.5% for Google to consider your page fast.

Again, if you get zero for FCP, there’s probably a Javascript problem you should take care of.

Fully Loaded Time

Fully loaded time takes it a step further from load time: it’s when the Document Complete or Window Loaded/Onload event has been achieved and the site is without network activity for more than 2 seconds. So basically, the difference between this metric and load time (Document Complete timing) is the additional time it takes for third-party code (analytics, affiliate code, ads, Facebook pixels) to load.

Keep in mind that in some cases, even a bit of marketing (a bit of third-party code) can make this value seem really high. In these cases, the result may be misleading.

LCP Timing (Render Mostly Finished)

Largest Contentful Paint (LCP), a Google Core Web Vitals metric, measures when the largest content element in the viewport becomes visible to the end user. In general, when the largest content element becomes visible, you can also consider the render to be nearly finished.

You should aim for LCP timing no longer than 4 seconds. LCP under 2.5 is considered to be fast by Google, while LCP over 4 seconds is considered slow – and slow is bad.

If you get a value of zero for LCP, there’s a good chance that there’s a Javascript or a 404 error hiding in your site code.

Cumulative Layout Shift (CLS)

Cumulative Layout Shift (CLS) considers the unexpected layout shift that takes place during the page loading process. This is when a user visits your website and, let’s say, wants to click on a link or video. However, just as they’re about to click, something in the page suddenly happens which shifts the entire layout. So, the visitor ends up clicking on something wrong, or loses their place in the text because the font changes and pushes the layout out of whack. This is often due to asynchronously loaded elements, a media file with unknown dimensions, or third-party code that either arrives late at the party or is doing its own thing, in terms of dynamics.

In any case, knowing the CLS value can help you get a clearer idea of how and whether the end user experiences a layout shift. Ideally, the CLS value should be under 0.1. If it’s over 0.25, there’s probably more layout shifts than desirable, and the user experience is considered to be poor.

Keep in mind that CLS is an important user-centered metric that shouldn’t be neglected.

Total Blocking Time (TBT)

The total blocking time (TBT) is the time that passes between FCP (First Contentful Paint) and the Time to Interactive (TTI). In other words, TBT measures the time between the beginning of the render and the moment when the website becomes fully interactive for the user. It’s the time in which the user can perceive the page as loading, but can’t interact with any elements, as the site won’t respond to their commands or input. During this time, the page is “blocked” – thus the name.

As with most of our other metrics, lower is better. Ideally, your TBT value should be under 300ms, or less than 0.3 seconds.

Time To Download On 1.6 Mb Internet Connection (Slow Speed)

It’s always a good idea to know how long it takes for your page to load on different-speed connections. We first measure how much time it takes for it to load on a slow connection of a 1.6 Mb (megabit) per second broadband connection or 3G mobile connection.

This metric, then, measures how fast your page would download on a slow connection based on its size (page weight). Note that this is also the download speed that Google takes into consideration when calculating mobile Pagespeed Insights Score. In other words, it’s important to make your page accessible to users with a slow connection, too.

Time To Download On 10 Mb Internet Connection (Moderate Speed)

This measures the time it takes for your page to download on a moderate speed internet connection, which we take to be 10 Megabit per second broadband connection or 4G mobile connection. While the value may be higher for slow connection, you should aim for a loading speed of under 3 seconds for moderate internet speed connections.

Time To Download On 25 Mb Internet Connection (Faster Speed)

Lastly, we measure how fast your page will download on a faster internet connection. A faster broadband internet connection or 4G mobile connection is considered to start at 25 Megabits per second. This value should usually be under 2 seconds, and ideally – less than 1 second.

Additional Analytics

Under our metrics and timings, we also provide you with some analytics about various aspects of your page and website. We’ll give you tips on what you can improve (like reducing page size, revising 3rd-party code), what you should look out for (warnings about what elements may be causing your site loading trouble), and what you’re already doing great (it’s not all criticism!).

We’re also launching a beta SEO SERP analyzer below the site analytics where we offer some suggestions on how you can try to improve your site’s online visibility and ranking.

Lastly, you can go over a visual overview of all your stats, including detailed metrics about elements on your page that are relevant in loading speed. A good lookover can help you polish up your code so nothing slows the page down.

Leave a Comment