The recent post about Sonys brand new horribly performing update to their Playstation Store kinda got me thinking, do people even care about the performance of what they produce these days?
I realize that dev.to at least does, that was always an important part of what Ben focused on, especially at the start.
It is also something I tend to spend at least some time on as well, and typically that means I can achieve nice PageSpeed scores. Sadly, one of the things I am working on is built on top of HubSpots CMS, and that in itself means its not really possible to get all the way =/
With that said though, do you care about PageSpeed? If Yes, care to show case some of your work? If No, why not? =)
Top comments (15)
For my two main projects, these are the scores (Left Mobile, Right Desktop).
These are pretty good scores! Another thing about testing your site's pagespeed (at least for me anyway) is that it is a great learning experience. Figuring out the stuff that is wrong helps you learn to do things better/more correctly. Same thing with HTML validation. I learned so much through deciphering all my problems/validation errors :)
Definitely. It makes you call some "best practices" into question at time even. And it does help you understand a little more about how the browser works as well.
It's important to remember that page speed does not always reflect real UX – it's just an automated tool that collects stats. The real UX comes from user perception of performance. For example, server side rendered applications that load >5 mb of javascript can feel really fast (if configured correctly) but will not win the award for the most performant site.
Also you should try Lighthouse, which does more check with new metrics – you can find in Chrome dev tools.
Here is mime result for omatsuri.app/ – I've lost some points on Github Pages caching strategy
Absolutely, that is ultimately what matters. If you load stuff the right way it can still be real nice. That is a great score as well : )
For groobe.io
Ever since I found lighthouse I have been always tried to improve it, it has been fun to do it though.
Major optimizations were created by using React Concurrent mode and prefetching the data, the site has a lot of images so I tried my best to make the loading experience great!
What I've learned is that clients like to pretend that pagespeed doesn't matter to them until they realize their rankings are tanking and get fewer clicks.
I'm working in ecommerce and we warned our clients several times that something they wanted would impact pagespeed (and rankings). Sometimes they'll listen, other times they'll call you a few months later and complain until you quote their mails ignoring the warnings.
I feel like gaming stores/launchers are especially atrocious, I never had a smooth UX in Origin regardless of hardware. Clicking stuff takes ages and animations lag, navigating back would take you back to the top of the list and not to your last position. And it's been like this for ages but they've recently announced a beta for a new launcher.
Uplays websites are in a similar boat, GOG and Blizzard seem to care more about UX (both launchers and websites).
Maybe they don't care because they know that the customer is forced to use the website if he wants to get a PS4/EA/Ubisoft-game, unlike in ecommerce where most products are available in several shops.
Its facinating with EA/Ubi etc given the rather extreme perf focus most game devs has as opposed to what their web/launcher devs seem to have.
I think that if your website needs more than 3 seconds to load, it's useless to have a website. This is certainly true for websites that are in the e-commerce world. Try opening Amazon and compare it to other e-commerce platform. The faster your content is available, the more likely the user is going to checkout the content.
For my website, I use Gatsby to optimize SEO and page-speed.
Here are the stats from lighthouse.
Google search ranking algos are starting to penalize sites that don't meet basic standards for #webperf core metrics.
Measure your site here:
web.dev/learn/
Good explanation here:
web.dev/user-centric-performance-m...
If you build sites, you need to know what these mean:
FCP, LCP, FID, TTI, TBT, CLS
Watch a vid about it, with all-around-studs Paul Irish and Elizabeth Sweeny:
youtube.com/watch?v=BjfYb8YiK1M
I care. Browser rendering time affects the user experience and ultimately SEO.
SEO has been a big reason for me as well (obviously User Experience is No 1), but for SEO its basically relative to other competing sites for the same keywords, and they are very often not scoring super well. So good for us I guess =)
What I mean is user experience via slow pagespeed affects SEO because google knows when people bounce from your slow site and then they gig you for it :)
As they should, there has been many a time when I've hit ghe back button when something isn't loading properly.
Just the other day poor UX also made me choose a competitor to deliver a product for $1200, so yeah, it has an impact for sure.
For Thomasbnt.fr