Gabriel Torres

I agree that the "mark as solved" link should not be part of post 1. So I got a solution to hide it.
Go to following template:
Forums -> front -> topics -> post
... and add the green code.
 

This is something that always bothered me. There used to be a plugin that fixed this, created by Matt Mecham (MTM Post Counter a.k.a. MTM Ye Olde Post Count Of Yesteryear), that is no longer available. I edited it to work with 4.5. See below. After installing it, you must clear the cache using the support tool. It is working like a charm here.
 
MTM_Post_Counter-4.5.xml

After researching about this, browsers have now native lazy loader. We only need to edit the templates to add:
loading="lazy" To all img tags.
I edited the following templates:
core > front > global
reactionBlurb
reactionLog
reactionOverview
reactionTabs
reputation
userPhoto
userPhotoFromData
This enabled lazy loading for user photos and reactions, and the "Defer offscreen images" warning from Google Lighthouse is now gone! 🙂
(I also made the same edit in my cms custom templates)

This is something that always bothered me. There used to be a plugin that fixed this, created by Matt Mecham (MTM Post Counter a.k.a. MTM Ye Olde Post Count Of Yesteryear), that is no longer available. I edited it to work with 4.5. See below. After installing it, you must clear the cache using the support tool. It is working like a charm here.
 
MTM_Post_Counter-4.5.xml

@xtechSo far, I am not seeing any improvement, as the main issue, which is the LCP, is still present. As mentioned in my original post, Google made a major change in its algorithm in the beginning of december, which is really hurting us. We used to get around 140k users per day from Google, since then we are getting around 115k. This is hurting our revenue tremendously. I am still working on this, though. Every couple of days I fix something here or there. I am very persistent and I will get to the bottom of this and will eventually fix this LCP issue! 🙂
In any case, this has been a good opportunity to fix a lot of other things that could be improved anyway! 🙂
 

@SeNioR- @Adlago @Mopar1973Man and everybody else...
Today I was able to make a really nice adjustment to our install here that improved our "Best Practices" score a lot:
Enjoy!

@AlexJ Please open your own topic; your question has nothing to do with my topic here.
 

@Morgin Got it. As mentioned above, I disabled it. Thank you very much for coming back and detailing the issue! 🙂
 

Thanks @Ryan Ashbrook for detailing what @Morgin mentioned. I will disable it "just in case".

@Adlago Many thanks for this tip, done that. I will investigate a little bit further every day, I made saveral small adjustments and fixes during this weekend. Let's wait and see how the metrics react to them.
I was just wondering why adding the font awesome preload at the end of the page and not at the start...
https://developer.mozilla.org/en-US/docs/Web/HTML/Preloading_content

It seems these FCP/LCP issues are being caused by the advertisement banners we have on our website. I made a few changes, but will continue investigating this further in the coming days. Thanks.

Fixed all that I could. There are still errors caused by:
(1) Cloudflare's Rocket Loader (A script element with a src attribute must not have a type attribute whose value is anything other than the empty string, a JavaScript MIME type, or module). I am not sure if there is anything we can do about those.
(2) Group Name Indicator plugin, which the developer has just fixed for us and we are waiting for approval at the marketplace.
I will take a closer look at the FCP/LCP metrics tomorrow. Pointers are welcome.
Thank you for your help! 🙂

Ops, forgot the most obvious... LOL https://www.clubedohardware.com.br
Many thanks! 🙂
 

Hi,
We've been badly hit by the Google December 2020 Core Update: our traffic comming from Google dropped 20% starting December 4th, 2020.
Investigating this further, I came across the Core Web Vitals, available at the Google Search Console, and our numbers are really bad there, especially for mobile. See screenshot below.
I am wondering if you guys are seeing the same (meaning this is something to be improved in the IPS software itself) or if this is caused by our custom modifications such as advertisement/banners.
Thanks in advance for any tips on this issue.

FCP - the problem is fixed with  Critical CSS.
For LCP, you can try preloading img, especially those img that slow down the mobile. Use for each img in your head
for mobile devices
<link rel="preload" media="(max-width: 979px)" href="url to your img" as="image">
desktop only
<link rel="preload" media="(min-width: 980px)" href="url to your img" as="image">
for everyone
<link rel="preload" media="all" href="url to your img" as="image"> Usually for mobile are a few img, incl. logo, for desktop you need to experimentally determine - whether all the img visible when loading the site, or only the img with the highest weight ...
Experiments give birth to the truth ...

@xtechSo far, I am not seeing any improvement, as the main issue, which is the LCP, is still present. As mentioned in my original post, Google made a major change in its algorithm in the beginning of december, which is really hurting us. We used to get around 140k users per day from Google, since then we are getting around 115k. This is hurting our revenue tremendously. I am still working on this, though. Every couple of days I fix something here or there. I am very persistent and I will get to the bottom of this and will eventually fix this LCP issue! 🙂
In any case, this has been a good opportunity to fix a lot of other things that could be improved anyway! 🙂
 

@xtechSo far, I am not seeing any improvement, as the main issue, which is the LCP, is still present. As mentioned in my original post, Google made a major change in its algorithm in the beginning of december, which is really hurting us. We used to get around 140k users per day from Google, since then we are getting around 115k. This is hurting our revenue tremendously. I am still working on this, though. Every couple of days I fix something here or there. I am very persistent and I will get to the bottom of this and will eventually fix this LCP issue! 🙂
In any case, this has been a good opportunity to fix a lot of other things that could be improved anyway! 🙂
 

They said that there is a SEO audit coming soon. They will probably take everything into consideration.

@Gabriel Torres  are you already seeing Google Analytics benefits of these improvements? I mean, how are the several optimizations you are making impacting your search engine traffic?
I believe that IPS should implement these improvements built in. Having successful communities is of their best interest, and SEO effectiveness should be a top priority.

Disappointed you didn't mark this as solution! 😆
On a serious note, I've always thought that IPS could develop the feature to the next stage.  "Marking as solution" could mean many things ... It could mean the post was helpful, the post offered alternate options, the post offered helpful hints, the post explained why the question was irrelevant to begin with, or the post addressed the specific issue.  
I've also found, in my community, that multiple members submitted various solutions.  All of them were partially valid.  
 

@SeNioR- @Adlago @Mopar1973Man and everybody else...
Today I was able to make a really nice adjustment to our install here that improved our "Best Practices" score a lot:
Enjoy!

Hello,
I am in the process of optimizing our install with the help with Google's Lighthouse, and trying to improve our score in that tool. 
Today I succesfuly worked on a issue that I suggest IPS to address in an upcoming release: multiple-resolution logo images.
Here is the warning that we got using lighthouse:

Following the suggestions decribed here: https://web.dev/serve-responsive-images/ and here: https://web.dev/codelab-density-descriptors/, I decided to create three logo images for our install at sizes 460x77 (1x), 920x154 (2x), and 1380x280 (3x). I uploaded them to a new folder (/uploads/logos/), and adjusted the HTML code at the core > front > global > logo template to:
<a href='{setting="base_url"}' id='elLogo' accesskey='1'><img src="{$logo}" srcset="/uploads/logos/logo-cdh-1x.png 1x, /uploads/logos/logo-cdh-2x.png 2x, /uploads/logos/logo-cdh-3x.png 3x" sizes="460px" alt='{setting="board_name" escape="true"}'></a> Note: I uploaded the 1x version as the logo in the template's settings page.
After this change, our logo looks much crispier and also we got rid of the above warning. In fact, our "Best Practices" score inscreased from 86 to 93! 🙂

I really hope that this helps other administrators who are looking into ways to improve their community's performance, and I hope IPS can make adjustments to fix the recommendations presented by Lighthouse to improve the overall score of our websites.
Enjoy! 🙂
Gabriel.

Hello,
I am in the process of optimizing our install with the help with Google's Lighthouse, and trying to improve our score in that tool. 
Today I succesfuly worked on a issue that I suggest IPS to address in an upcoming release: multiple-resolution logo images.
Here is the warning that we got using lighthouse:

Following the suggestions decribed here: https://web.dev/serve-responsive-images/ and here: https://web.dev/codelab-density-descriptors/, I decided to create three logo images for our install at sizes 460x77 (1x), 920x154 (2x), and 1380x280 (3x). I uploaded them to a new folder (/uploads/logos/), and adjusted the HTML code at the core > front > global > logo template to:
<a href='{setting="base_url"}' id='elLogo' accesskey='1'><img src="{$logo}" srcset="/uploads/logos/logo-cdh-1x.png 1x, /uploads/logos/logo-cdh-2x.png 2x, /uploads/logos/logo-cdh-3x.png 3x" sizes="460px" alt='{setting="board_name" escape="true"}'></a> Note: I uploaded the 1x version as the logo in the template's settings page.
After this change, our logo looks much crispier and also we got rid of the above warning. In fact, our "Best Practices" score inscreased from 86 to 93! 🙂

I really hope that this helps other administrators who are looking into ways to improve their community's performance, and I hope IPS can make adjustments to fix the recommendations presented by Lighthouse to improve the overall score of our websites.
Enjoy! 🙂
Gabriel.

Interesting idea, but what of the member who to circumvent this replies a first time... waits a bit, replies a second time, and then marks that second post as a solution to achieve the same goal (attention seeking behavior)?
I can see a case for thread author not being able to provide their own solution (require the solution be from a member other than the thread author), yet I'd suggest that this would be a moderation/policy issue. I'm not sure I can think of a technical solution that isn't abusable.

Hi there,
We are seeing some users here in our community who are creating topics and immediately marking their single post as solution. We guess they are doing this only to get attention.
So my suggestion is to improve this feature by not allowing the topic author to mark the first post in a topic as solution. The user would still be able to mark other posts (including his own) as solution in the topics he started, just not the first one.
See if that makes sense.
Thanks.