PageSpeed Insights - Score

[Phil Këvin]

It would be great if they could work on the optimization. This negatively affects SEO, so I would like invision to prioritize this optimization enhancement, especially in the Mobile part.

PageSpeed Test & Score for Invision Alpha

• Home Page
• Topic/Forum

PageSpeed Test &Score for InvisionCommunity

• Home Page
• Topic/Forum

Those who have read this topic, please share the score of your website.
This is the score of my site. (I know that my site needs a good optimization.)

Mobile

 

Desktop

Edited May 28, 2020 by Phil Këvin

[Feneroin]

Please be aware that this test site is running in 'development mode' so it is automatically updated with the latest fixes throughout the day. This means it has to work extra hard on each click as there are no caches, pre-built languages or templates to use, so it will be a lot slower than a production version. So please don't worry about it being a touch slow, and definitely don't try and run Page Speed analysis tools on the alpha site!

Edited May 28, 2020 by Feneroin

[Phil Këvin]

20 minutes ago, Feneroin said:

 

Obviously it is in beta, but I leave it there in case they happen to forget.

[sadams101]

Google just increased their requirements. Now serving JS on pages that it isn't needed is hurting speed scores badly. Test again and you will see this. I was scoring 80 on mobile, and 97 on desktop, until a few days ago when this new requirement came out. Now I score ~48. I'm trying to find out now if I can have a plugin made to serve only JS that is needed per page/module.

You may want to reach out to @Adlago who has helped me greatly to optimize my site. If I get a plugin made for this I'll let you know.

[Phil Këvin]

11 hours ago, sadams101 said:

Google just increased their requirements. Now serving JS on pages that it isn't needed is hurting speed scores badly. Test again and you will see this. I was scoring 80 on mobile, and 97 on desktop, until a few days ago when this new requirement came out. Now I score ~48. I'm trying to find out now if I can have a plugin made to serve only JS that is needed per page/module.

You may want to reach out to @Adlago who has helped me greatly to optimize my site. If I get a plugin made for this I'll let you know.

Hello,

Very kind of you, thanks for the information 🙂

[Sheffielder]

On 5/29/2020 at 6:15 PM, sadams101 said:

Google just increased their requirements. Now serving JS on pages that it isn't needed is hurting speed scores badly. Test again and you will see this. I was scoring 80 on mobile, and 97 on desktop, until a few days ago when this new requirement came out. Now I score ~48. I'm trying to find out now if I can have a plugin made to serve only JS that is needed per page/module.

You may want to reach out to @Adlago who has helped me greatly to optimize my site. If I get a plugin made for this I'll let you know.

 

Well that's just terrible news

Are Invision aware of this and building it into the update?

Wow just scored 29 on my test

That's the lowest I've ever scored by miles

[marklcfc]

8 minutes ago, Sheffielder said:

Are Invision aware of this and building it into the update?

Probably, doubt it

Edited June 1, 2020 by marklcfc

[aXenDev]

I see these topics probably several times a week.
They wrote that they strive for the best performance, but no system is without flaws and are not able to draw those 100% compliance.

Even on tests, google pages are not perfect.

[Sheffielder]

1 hour ago, aXen | 1s2k said:

I see these topics probably several times a week.
They wrote that they strive for the best performance, but no system is without flaws and are not able to draw those 100% compliance.

Even on tests, google pages are not perfect.

 

Im talking about the new change from Google though 

[aXenDev]

1 hour ago, Sheffielder said:

Im talking about the new change from Google though 

I know

[sadams101]

Some speed issues I've pointed out at more than 2 years ago, both on this forum and in support tickets, are still present. This is why I've gone off on my own path and have worked with people who have overcome most of them. I was a solid 80 on mobile before this new update. This new JS requirement will be harder to deal with, but I'm working with a programmer now to see what can be done. If I find anything out I'll post it here.

[sadams101]

It has taken me ~3 months since upgrading to 4.5 (working with @Adlago) to recover my site speed, and eliminate the dozens of W3 html errors inherent in the standard version of IPB, and I won't be stopping until I'm in the 90% area for mobile (I'm currently at 97-98% for desktop). I'm just shocked that these issue continue in the latest versions, and in some ways were actually worse and harder to deal with in 4.5 than in 4.4.

 

[Apfelstrudel]

Thanks for the info. I have the same problem.

Could you please share some insights of the major changes?

[Thomas P]

11 hours ago, sadams101 said:

It has taken me ~3 months since upgrading to 4.5 (working with @Adlago) to recover my site speed, and eliminate the dozens of W3 html errors inherent in the standard version of IPB, and I won't be stopping until I'm in the 90% area for mobile (I'm currently at 97-98% for desktop). I'm just shocked that these issue continue in the latest versions, and in some ways were actually worse and harder to deal with in 4.5 than in 4.4.

One difference from 4.4 to 4.5 was that pagespeed was improved a lot.

Did you try to use the default theme? Try your test installation forum with default theme and let pagespeed visit it.

In March this was our result with the default theme and no ads:

Edited May 26, 2021 by Thomas P

[sadams101]

For the record I am using the default skin, and not a custom skin. I do have a custom footer, and a custom menu, as well as other plugins.

I think we all know that the mobile site speed scores are what affect rankings the most now. While your home page score looks pretty ok, not great, let's face it, 90%+ of your entire site, like mine, consists of topics, so those are the most important scores for Google site speed ranking. Take a look at one of your thread page scores for mobile, and this is exactly what I saw after upgrading to 4.5, so I'm not criticizing your site in any way:

 

Even the makers of this software, who are using the next version of this software 4.6, are still doing horrible for site speed...see one of their threads below:

The architecture of the site needs to be changed because the CSS loads EVERYTHING for the mobile site, and EVERYTHING for the desktop site. There are many other issues that are causing this, but IPS has promised faster speed scores, but 4.5 and 4.6 are both terrible.

PS - I won't even go into the number of W3 html errors in the software, which also affect ranking. Most are easily fixed as I've done...I reported these to IPS as well, but it's always going to be fixed in their next version (4.6 has even more errors!). These also affect rankings.

https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.mcseboard.de%2Ftopic%2F220055-windows-10-kein-reconnect-bei-netzlaufwerk%2F 

Edited May 26, 2021 by sadams101

[Jimi Wikman]

May I ask why you care so much about these tests?

Are you trying to grab an extra position or two or do you just want to have things in the green?

Not trying to mock you or anything, just wondering what the end goal is for you to increase completion of these tests because I am always curious to learn what drives these type of completion chases.

[Ibai]

1 hour ago, Jimi Wikman said:

May I ask why you care so much about these tests?

Are you trying to grab an extra position or two or do you just want to have things in the green?

Not trying to mock you or anything, just wondering what the end goal is for you to increase completion of these tests because I am always curious to learn what drives these type of completion chases.

Maybe SEO recommendations? Page Speed?

I agree there are several things that might be improved in IPS.

[Jimi Wikman]

7 minutes ago, Ibai said:

Maybe SEO recommendations? Page Speed?

I agree there are several things that might be improved in IPS.

Pagespeed is always nice, but unless people complain, or you see loss of sales, how do you measure when you succeed or what the value is? Pagespeed comes with a cost often and sometimes that cost will be significantly higher than the benefit.

SEO wise I don't see many professional SEO agencies focus much on that unless you are competing in the top 25 or so for specific keywords? So putting time and effort into shaving off 2 seconds in mobile view when you are at placement 600 is usually nice, but not really where focus should be?

This is why I am always curious 🙂

[sadams101]

Are you really asking me to explain why SEO is important? You can ignore Google's ranking metrics at your own peril, but for a company whose sole product is making software for others, when they ignore this it affects everyone who uses their software, and it affects everyone's page rank in Google. Maybe you should ask yourself why you have any Web sites on the Internet at all?

Edited May 26, 2021 by sadams101

[sadams101]

14 hours ago, Apfelstrudel said:

Thanks for the info. I have the same problem.

Could you please share some insights of the major changes?

If you are serious about this I recommend that you contact @Adlago, as he's my speed consultant.

For what it is worth, IPS could fairly easily implement all that I've done into any current of future versions of their software, but it's apparently not a concern for them.

[Jimi Wikman]

10 hours ago, sadams101 said:

Are you really asking me to explain why SEO is important? You can ignore Google's ranking metrics at your own peril, but for a company whose sole product is making software for others, when they ignore this it affects everyone who uses their software, and it affects everyone's page rank in Google. Maybe you should ask yourself why you have any Web sites on the Internet at all?

Not at all, I am just wondering if you are competing at the level where that actually will make or break anything for you? Onsite SEO is of course best practice that your software support, especially if you have a commerce application and page speed should be acceptable of course. Having the fastest site though will not give you the top spot in rankings, not even by a long shot. It is an edge against competitors only if their offsite SEO is on equal ground to yours.

I have done quite a bit of work with CRO and I even teach Neurowebdesign, so I know the benefit of sales for even fractions of a second. So I understand page speed for sales, but for community sites speed is not as important (no, I don't mean slow sites are accepted, I mean very few people will care if the site loads in 1.1 second or 1.0 seconds, or even 2.5 to 2.0 seconds).

For rankings, I see a lot of people put in a  ton of work and money to go from a ranking of 12000 to 600 for example and then get sad when they realize that position 600 is pretty much the same as 12000 because no one will see it anyway.

Is your focus with the pagespeed effort to convert more members, rank on certain keywords or do you have a conversion funnel for sales that I could not see on your site as a guest?

Also, if you don't mind, the effort you put in how many positions did that earn you on your top keyword(s)? It would be really interesting to see the result of your hard work beyond just having a very snappy and fast website (looks great btw). I think that can help others when they consider putting the time, cost and effort into optimizing for speed.

 

...and no, I never ask myself why I have websites 🙂
I sometimes ask my clients, but I have websites for two reasons: I like to build things and I like to write. That is why most websites exist, which is the difference between a business oriented website and a hobby website for me.

[sadams101]

I've seen dramatic improvement in both site traffic and keyword rank by making SEO changes over the 25 years my site's been online. I've seen huge changes up and down depending on issues I've fixed or new issues that surfaced. For example, I saw a HUGE drop in my rankings that began on the very day that I upgraded my site from 4.4 to 4.5 on March 26 (but I ran some pre upgrade tests March 2-10 which also show), due to:
1) Site speed issues which I had mostly resolved in 4.4;
2) Site errors;

In fact, during the exact time period covered by the Google Core Web Vitals below, my site dropped from a global rank of 228K to 315K. a 27% drop! Did I see this drop in traffic, hell yes I did, I lost over 100K unique visitors per month, and the number of accounts created, posts made, etc. all dropped by over 30%.

What has changed since then? I fixed all the HTML errors--there are literally dozens per page (for which there is absolutely no excuse), and got my mobile page speed scoring above 70% average, now pushing closer to 80%...that's it.

Before this I paid for plugins to also fix all the unexplained, unnecessary 301 redirects that are inherent in the blocks, sharing links, page pagination, etc. Internal 301 redirects are just toxic for SEO...no two ways about it. Any SEO person worth their salt would tell you this. I had 10-20 301 redirects per page average before using plugins created by @DawPi fixed them (no downsides whatsoever...all links function perfectly, just no redirects--all the silly ?do=find stuff).

Each major SEO fix I've done has shown clearly visible results in traffic, rank, etc. I've reported all of them on this forum, and all to IPB directly, and I get the usual replies...the usual crew jump in to tell me none of this really matters, it's no big deal, etc. Great...maybe for your site, but not mine. You're either in the software business to make the best software for your customers, and SEO is now a huge part of that, or you're not. IPS has the power to implement all the changes into the software that I paid a few thousand dollars to fix these issues, why not do this for everyone and help everyone's SEO and rank? Instead we get badges...great! Badges get me more members...NOT!

The bottom Alexa chart tracks my recovery, as @Adlago has rolled out all the site speed fixes over the last couple of months.

 

Edited May 27, 2021 by sadams101

[Jimi Wikman]

Very interesting and a nice flashback seeing Alexa rank 🙂
I though that died out years ago, but there you go 🙂

Ok, so you see a drop when your Alexa rank drops, that would be because some of the keywords you rank on drop down below the 25-30 top ranking I assume? The graph looks a bit odd though because it looks like you had a degrading situation for a while and then it jumped up? Are you sure you weren't in a Google filter during that time because it looks like you got back up to roughly the same levels, but still slightly lover from the previous peak?

May I ask what speed difference you experienced on desktop and mobile after the upgrade?

It seems strange that you have so many errors though. I have 6 errors on my site when checking and it is all my fault (hence very easy to fix) and this forum have 1 error (missing ALT). Nothing that should be a cause for any concern unless you are aiming for an AAA WCAG rating?

I take it that my site that has no SEO or speed optimization also have these redirects, so would you mind pointing me to one so I see what kind of mayhem it's causing?

Thank you for sharing, I find these things fascinating and very interesting!

 

[Thomas P]

On 5/26/2021 at 7:09 PM, sadams101 said:

While your home page score looks pretty ok, not great, let's face it, 90%+ of your entire site, like mine, consists of topics, so those are the most important scores for Google site speed ranking. Take a look at one of your thread page scores for mobile, and this is exactly what I saw after upgrading to 4.5, so I'm not criticizing your site in any way

I take it as a compliment 😉
To be fair: As I said we measure without Ads, this is content we could not influence. Plus topics consistent of User Generated Content (UGC) and in the sample you picked we have several banners and a screenshot. Most of the posts have >50-60%

Secondly Pagespeed does not work worldwide, they are measuring from across the pond from their HQ.
What about your GTMetrix numbers?
Log in there, use a server next to you and please tell me what you see.
This is what I see:

To achieve good numbers PHP Workers, GZIP/Brotli, Redis and several other components and configs have to be well tweaked. Just saying that code is not all.

On 5/26/2021 at 7:09 PM, sadams101 said:

The architecture of the site needs to be changed because the CSS loads EVERYTHING for the mobile site, and EVERYTHING for the desktop site. There are many other issues that are causing this, but IPS has promised faster speed scores, but 4.5 and 4.6 are both terrible.

I see that IPS is heavily improving their code performance and seo-wise from release to release, undoubtedly they do a good job compared to other comparable platforms.
There is always room for improvement, but are you sure that 4.5 and 4.6 -which is still beta- is "terrible".

How does a Wordpress perform on pagespeed and GTMetrix on your server?

 

 

 

 

[sadams101]

 

@Jimi Wikman, yes, Alexa Rank is still important, especially after Amazon bought them. It's probably more important now than ever, especially if you ever try to sell your Web site. The 90 day time frame covered in the Alexa rankings pretty much matches up with the period of my March upgrade, and it shows the decline and recovery from the worst of the site speed and html errors (IPS must have taken my post seriously about the 120+ warnings per topic page in this forum, as they have now fixed most of them). But this isn't the first time I've experienced this, as I've run upgrades in the past, and have had similar experiences. 

But seriously, if you don't believe me about this, perhaps check with SEO companies:
https://cognitiveseo.com/blog/22865/page-speed-seo/#:~:text=The simple answer is that,rate and reducing dwell time.
or with Google directly (this started in 2018, and as you may know, they typically add more weight to this as time goes by):
https://developers.google.com/web/updates/2018/07/search-ads-speed

You mention that "they realize that position 600 is pretty much the same as 12000 because no one will see it anyway," but if they have a slow web site they will never have a shot at a first or second page search ranking no matter how great their content...they will be shut out by the algorithm. The only chance you have at ever getting first page rankings on any key words is by jumping through Google's hoops, and they have even created tools to help you do this, and this is the tool I focus on because it's what Googlebot is based on: 
https://developers.google.com/speed/pagespeed/insights/

I agree that it's hard for individuals to try to take software like IPB's and optimize it for speed, but what I am saying here is that I should not have to do this at all--IPB should do this. By not doing it they are pretty much guaranteeing that anyone using their software will never achieve high Google rankings, no matter how great the site, and how great the content.

@Thomas P I don't have a paid GTmetrix account so I can only run desktop free tests there, however, to me the only speed tests that matter are Google's, which uses Lighthouse as their test engine. I did just compare both our sites, and again, I never use my home page to test, and always use a topic page, as that makes up the vast majority of my site, and yours as well. Even with all the ads loading on my site I clearly have a much better score from the comparison test I just ran below, but again, my focus is the horrible mobile performance of IPB, which will hurt your rankings far more than the desktop performance. I suspect that if we compare the same two mobile pages in GTMetrix you would get an F and I would get a B or C: