Google new site audit tool

[gpietersz]

Its does what pagespeed does and a lot more: https://web.dev/measure/

I have tried it out quickly on a few sites and looks like it will be both useful when fixing problems and it presents problems in a way that is relatively easy to understand even if you do not understand the technology (especially the five headline scores at the top).

 

[fisicx]

Had a bit of a play and I get different results depending on which tools I use.

But it's still a useful tool - if of course you know how to fix the things it's reporting!

 

[UKSBD]

is it just a trimmed down version of developer tools under more tools in chrome?

 

[WebDesires]

Firstly, I appreciate you just found it but it certainly isnt new... Has existed now almost a year.

Secondly, it is just a trimmed down version of Lighthouse. While useful you can do tests quicker and with better results from the developer tools either in Chrome or Opera (or any other Chrome based browser).

 

[gpietersz]

is it just a trimmed down version of developer tools under more tools in chrome?

Yes and no. It shows some numbers I do not see in the latter (cumulative layout shift) .

It also warns about up an old JS library with known XSS issues on one site I tried it on which Chrome developer tools does not.

It also gives you the full report as a shareable link, and its been tied into Search Console (I think that happened recently) and I find it a lot easier to read.

Secondly, it is just a trimmed down version of Lighthouse.

it is Lighthouse - it says so if you click on view report.

Why are results from Chrome developer tools better?

 

[UKSBD]

It also warns about up an old JS library with known XSS issues on one site I tried it on which Chrome developer tools does not.

It should show that in Best Practices

It also gives you the full report as a shareable link, and its been tied into Search Console (I think that happened recently) and I find it a lot easier to read.

Developer tools creates a html file you can upload and link to

https://www.uksbd.co.uk/lightdemo.html

 

[gpietersz]

It should show that in Best Practices

It did not in this case.

Developer tools creates a html file you can upload and link to

I did not know that and tis useful to know - but its not as convenient as the link that is right there and having it hosted by them.

 

[WebDesires]

Why are results from Chrome developer tools better?

a multitude of factors really, location of the testing tool, for example if the tool is running from USA and your site is in UK then the timings will be worse. Also network speed, ie if many people are running the tool at once the tool will be fighting other network traffic and thus tool will report timings worse - which is the main reason I do not take the results as perfect.

Then you have version, the one in your browser will more likely be most up-to date.

I believe the lighthouse library runs its tests using a real browser, so you also have the issue that it may not be using the same browser as you, of course your browser will be chromium but different builds behave slightly differently, also they behave differently depending on OS too, so you might be running from a Chrome in Windows and the tool Chrome in Linux.

Then you have plugins which affect lighthouse results, what I do is have a "silo'd" version of chrome that has no plugins whatsoever where I know then that plugins cannot affect results...

Dont forget also that if you are blocking for example google ads and your site has google ads then you will get a much better report from lighthouse because they will also be blocked when running the tool in your browser, this might be obvious but just in case you didn't realise.

All factors to keep in mind. And the hosted tool gives you no information on whats potentially blocked, where the tool is running from, what hardware or how congested it is. I just find more control and more comfort in knowing where im running it and under what factors.

 

[WebDesires]

here is the results from one of our clients sites, we have spent a lot of time optimizing, improving and crafting a solid SEO approach: https://lighthouse-dot-webdotdevsite.appspot.com//lh/html?url=https://www.rainvac.com

 

[gpietersz]

a multitude of factors really, location of the testing tool, for example if the tool is running from USA and your site is in UK then the timings will be worse. Also network speed, ie if many people are running the tool at once the tool will be fighting other network traffic and thus tool will report timings worse - which is the main reason I do not take the results as perfect.

All of those are also reasons to use the version on the site: for example Google might have better connectivity than you do, or you might be in the UK and want to see results from the US, etc.

Then you have version, the one in your browser will more likely be most up-to date.

I do not understand what you mean by that. Version of what?

I believe the lighthouse library runs its tests using a real browser, so you also have the issue that it may not be using the same browser as you, of course your browser will be chromium but different builds behave slightly differently, also they behave differently depending on OS too, so you might be running from a Chrome in Windows and the tool Chrome in Linux.

So why would your browser be better than theirs. This and the location issue look like reasons to try both or, using tools that will run from multiple locations.

Then you have plugins which affect lighthouse results, what I do is have a "silo'd" version of chrome that has no plugins whatsoever where I know then that plugins cannot affect results

I am pretty sure Google will not be rendering with a broswer that has a load of plugins.

I just find more control and more comfort in knowing where im running it and under what factors.

I agree to an extent, but how much control do you have over your ISP?

 

[fisicx]

here is the results from one of our clients sites, we have spent a lot of time optimizing, improving and crafting a solid SEO approach...

I've run test on some of my site and they score pretty poorly. But as the sites rank well and I'm getting very targeted traffic from around the world I'm not too worried.

 

[UKSBD]

Google have announced that from next year Core Web Vitals (the things these tools tend to measure) will be part of the search rankings algorithms

Great content will always be key, but how well a page performs, how well Google can determine if it provides good user engagement will give a little boost.

"A good page experience doesn’t override having great, relevant content. However, in cases where there are multiple pages that have similar content, page experience becomes much more important for visibility in Search."

source: https://webmasters.googleblog.com/2020/05/evaluating-page-experience.html

 

[fisicx]

To be honest, there is nothing new about this. All Google are doing is making it a bit more public.

The visual stability bit is interesting, it may make a lot of sites with page builders struggle because of how the page renders. You often see pages where some bits load and slowly fill with content and all the scripts kick in. It may also affect those sites with sliders and animations. Something sensible developers have known about for a while.

 

[NickGrogan]

It seems happy with some fair basic things. Interestingly it recommends removing google analytics to improve performance.

https://lighthouse-dot-webdotdevsite.appspot.com//lh/html?url=https://www.xn--courtiernergie-ikb.fr/

 

[Darren_Ssc]

It seems happy with some fair basic things. Interestingly it recommends removing google analytics to improve performance.

Google fonts is often another culprit although one can get carried away when going down the route of ultimate optimisation in order to get some great arbitrary score at the expense of a functioning website that's actually going to convert visitors into buyers.

 

[WebDesires]

I've run test on some of my site and they score pretty poorly. But as the sites rank well and I'm getting very targeted traffic from around the world I'm not too worried.

I would start to get worried, it is becoming more and more important, you dont want your business to slowly fizzle out or even worse get slapped one day.

To be honest, there is nothing new about this. All Google are doing is making it a bit more public.

The visual stability bit is interesting, it may make a lot of sites with page builders struggle because of how the page renders. You often see pages where some bits load and slowly fill with content and all the scripts kick in. It may also affect those sites with sliders and animations. Something sensible developers have known about for a while.

You are right, they have been slowly increasing importance on speed for quite a while, but it was a bit of both tweaking their metrix and also the ranking power, this has had the biggest issue that I've had which is the scoring tool has been getting more and more stringent and its been getting harder and harder to get the results you see from our main client above.

Ofcourse like i said above, this is getting more and more important, Google are pushing faster and faster the importance of speed and usability as a ranking factor.

So i recommend you start taking notice and improving your scores as much as you can, when you can.

 

[Darren_Ssc]

... its been getting harder and harder to get the results you see from our main client above.

Have you introduced your client to such tools? You may be making a rod for your own back if you have?

 

[WebDesires]

It seems happy with some fair basic things. Interestingly it recommends removing google analytics to improve performance.

https://lighthouse-dot-webdotdevsite.appspot.com//lh/html?url=https://www.xn--courtiernergie-ikb.fr/

Yes all of googles services are massive site hogs, the biggest recent example is Recaptcha V3, it causes such a bottleneck that it adds 3 seconds onto execution and even worse on render etc. We had to remove it from all our clients because of its huge detrimental impact on our clients sites - and we have seen HUGE boosts in conversions, sales and rankings within google itself.

So its definately worth the effort, remove as much SHIT as possible. Dont bother with analytics, its useless anyway. The stats from Webmaster Tools is good enough and directly from search instead of your site.

 

[WebDesires]

Have you introduced your client to such tools? You may be making a rod for your own back if you have?

Unfortunately yes, but just explaining how it all works suffices in lowing the tempers of poor scores. And its always great to hit the high scores eventually and prove your worth until the next update to the algorithm

 

[Darren_Ssc]

Unfortunately yes, but just explaining how it all works suffices in lowing the tempers of poor scores. And its always great to hit the high scores eventually and prove your worth until the next update to the algorithm

You could always run a few tests against some popular eCommerce sites and demonstrate to your client that they are x number of points better than amazon, etc.

 

[Richard Ben]

I am not very satisfied with the tool you have given me. Overall, it has to provide more convenience.

 

[ukbf0001]

Useful tool - I tend to use a combination of tools - gtmetrix , pingdom , page insights and website auditor

From there you can tweak the onsite seo as well as the other things like speed, FCB, image issues and all the other stuff. I wouldnt obsess too much though - having content that converts is the No1 priority. Also having a play with rankmath at the moment - but never have the time to set aside a couple of hours to setup some tests

 

[Aleksander_Gramm]

I would start to get worried, it is becoming more and more important, you dont want your business to slowly fizzle out or even worse get slapped one day.

You don't have to worry. Google uses 200+ rank factors (the content isn't key ranking factor but it is important because modern search engines use a semantic search).
Google doesn't use the same weight of ranking factor for different websites.
The main goal of Google to get the slow website owners (about 10-15% of websites) to improve their websites.
In other words if you get 100 for your website it doesn't boost a rank.
Also (IMHO) Google wants that website owners optimized JS, CSS, images. It all about skills of web developers. Many people get money to create spaghetti code. But, as we all know, it is a cheap way to create a website.
Google can't measure these metrics so they created own methods to measure "a website speed". Googlers recognize that it isn't good way. So, it is the minor ranking factor.

Also you can read this post https://www.seroundtable.com/google-lighthouse-seo-score-29730.html