4

From a speed POV using Googles fonts will normally be faster as (a) they will be cached and lots of sites use them and (b) Google will effectively be using a CDN to provide high speed results. There are minor downsides - Doing it this way is another way that Google can track your viewers and you are reliant on Google not removing this functionality. These ...


4

Google compress their fonts for faster download and if you are using HTTP/1.1 then downloading the fonts from this seperate domain could also result in a performance improvement. Whether Google or local, the fonts will be cached by the user's browser after the first visit, so from then until the files expire the same browser won't need to download the font ...


4

It will be field data. In their announcement of the algorithm update, Google are clear about using "real-world, user-centered metrics". By definition, lab measurements are not real-world, and what I see in my lab might not be the same as what you see in yours. Google themselves are clear on this in their documentation for speed tools: lab data &...


4

Chrome's incognito window uses a fresh cache, so you could say that Chrome has two caches - a "normal" cache and an "incognito" cache. When you close ALL incognito windows and tabs, the incognito cache gets cleared. With default Chrome settings, the normal cache persists across windows and browsing sessions. It takes extra time for your ...


3

One thing that others have not mentioned yet is competitor speed metrics. If you score an 86/100 and you have a competitor site that is very similar to yours (i.e. similar backlink profile and on-page signals) and they are scoring 96/100, then they probably have a good chance of ranking higher. Now, that is a made up hypothetical situation because there are ...


3

If you aren't able to detect the CLS issue using lab tools, then the issue is likely the results of user interactions or device/network combinations that you might be aware are common. I wrote a guide, Debugging Web Vitals in the field that explains how to use your analytics tool to detect what could be causing the CLS issue with your field data. I also gave ...


3

Is Difference Between German and English Versions Possible? Yes, it is possible. There can be several reasons. For example, most visitors from the US may use certain devices, browsers, etc., that German visitors may use sparingly. Server location, DNS speed, etc., may play a part. You have to segment the visitors from the US and Germany and profile them to ...


2

I think there are two options you have, the first would be trying to add the defer attribute to each script. It might work in conjunction with async to make sure the core info of your site (html and css) is loaded first. I believe (don't quote me on this though) that defer essentially reproduces what you proposed in putting the code in the footer - waiting ...


2

For the most part I think that compressing the images with higher JPG compression should suffice. I say this because we do know that Google cares a lot about speed - and the reason they are advising you of the sizes of your images is because they transfer unnecessary data and slow down your connection. I don't have an authoritative source, but according to ...


2

Yes, that is cloaking. Googlebot should see the exact same page that users see. Google understands javascript and you are denying Google a chance to process the same to determine if the scripts modify the content or not. While there can be variations based on geolocations or even mobile devices, any variation aimed specifically at Googlebot is risky. Below ...


2

Having the image URL change to the CDN's domain is problematic for image SEO. You say that an example image URL is now https://d33wubrfki0l68.cloudfront.net/1cc7c854a3e57dd05ecaf7f6153d5736820f3d76/58b8a/img/ingo-steinke-portrait-2020.jpg while the original URL on your site is https://www.ingo-steinke.de/img/ingo-steinke-portrait-2020.jpg The image is no ...


2

I'll say it boldly: Text-to-HTML ratio is a bad metric that has no place in SEO. If you're using Text-to-HTML ratio as a "warning flag" to pick out pages to take a closer look at, that's one thing. But it's not correct to say that the metric reliably points out "bad pages" - there are plenty of valid reasons a page could have low Text-to-...


2

If the 86/100 is the mobile score, the desktop score is likely ~95-100. I have many sites that score like this, and have no problems ranking content on either desktop or mobile. This will not hinder pages on your site from ranking. However, sinces Google's recent Page Experience Update, it does mean that you may lose a tie breaker where the decision is ...


2

Questions I'm trying to answer: are these visitors using a specific device/browser type? Note that Google's field data is coming from Chrome (and then only users whose privacy settings allow for it), so for what is shown in PSI, CrUX etc. there is no question of what browser was used. This is an important point because it is likely not a representative ...


1

It is a little strange to need a very specific tool, because there are other ones doing the same thing (displaying an histogram). I found that one https://golangrepo.com/repo/six-ddc-plow-go-testing-frameworks that seems to show kind of the same histogram as your example but not 100%. Difficult to see if it can be just a display difference or version ...


1

A good score in Google PageSpeed is good for the following reasons: Reduce the bounce rate. Increase the returning visitors. Better SEO positions on desktop and mobile. By the contrary, a slow site means: Visitors don't wait the page load and close before reading. Visitors will not return to your site for this bad experiencie. Bad SEO positions on desktop ...


1

Requesting an external resource on a website or application incurs several round-trips before the browser can actually start to download the resource. These round-trips include the DNS lookup, TCP handshake, and TLS negotiation. Your problem is that video is blocking the browser's main thread. It is "renderblocking". To add to Stephen's answer, if ...


1

Rather than use the default embed code from Youtube, you can truly lazy load the video: Put just a placeholder image in the HTML Run JavaScript onclick on that image that replaces it with the Youtube iframe and plays the video. The Youtube JS API fully supports this. Here is the code to make it happen from ak85's answer to this question on StackOverflow: ...


1

You could probably find some decent stats on this, but I think you've basically arrived at the right answers yourself. Yes, SEO is anecdotal, and you have the right idea about CDNs. I think it makes no difference to Google if you're properly redirecting and following best practices on that. Search for something in Google Images. Depending on search term you'...


1

Google measures core web vitals via CrUX reports - https://developers.google.com/web/tools/chrome-user-experience-report/ The metrics are measured in real world when real visitors visit your web page. These real world metrics aren't going to change by removing some jQuery libraries for lighthouse UA. Those core metrics are the ones that are going to ...


1

The lighthouse CLS score is calculated based on the current viewport it is being tested in. For example, if you test your site on a 1366px monitor and then on a 1920px monitor, the scores may be different (depending on your code quality). So if you're seeing a bad CLS score from the collaborative data on the Page Speed Insights tool, or in the Google search ...


1

"It depends" - On the functionality of the site and the tradeoffs one is willing to make. Pre-rendering is indeed very useful for static, high volume pages or static pages where cpu is very constrained - but you give up an awful lot to get there, including a lot of user interactivity and harder maintenance. Computers are unbelievably fast (and some ...


1

Is this to be expected and typical? In short, yes. Google's extensive documentation on Core Web Vitals, and specifically on cumulative layout shift, are explicit on this point: Caution: Lab tools typically load pages in a synthetic environment and are thus only able to measure layout shifts that occur during page load. As a result, CLS values reported by ...


1

Google Tag Manager makes easy and quick to set up Google Analytics goals. If you need tracking only page views, link clicks and scroll depth, you can do it without GTM. How to add analytics.js to your site or add alternative async tag How to set up Universal Analytics events How to capture outbound links


1

I faced the same issue. To solve it, I installed W3 Total Cache and also used Cloudflare. Now there is no issue at my website.


Only top voted, non community-wiki answers of a minimum length are eligible