I just wrote about swapping out MUI/Emotion styling for TailwindCSS and the PageSpeed benefits, but if you don't have other technical motivations, is it worth doing all that work just to get those speed benefits?
Of course, page speed matters, but...
This is not one of those black and white, it matters or it doesn't, things. In addition to (potentially) helping with your Google rankings, it just generally makes your user experience better.
Google came out about 10 years ago and said as much. It was one of those momentous algorithm changes that threw the snakeoil, sorry, SEO industry into spasms. Of course, no one likes sitting there waiting for their page to load. Humans increasingly have the patience of gnats.
But then consider this... would you rather have a fast page or a good page? The question doesn't really make much sense. An instant page of crap is still a page of crap. The more the rules are understood, the more the system is gamed, and the less meaning those metrics begin to have.
Content quality matters, too. And user experience. And messaging. And branding.
When you stop for a minute to think about it, it wouldn't make sense for Google to prioritize a fast page over a "good" page. The only think that makes any sense is for them to figure out a way to tip the scales to the faster of two relatively equally relevant pages.
So, my advice is to give your site a run at the PageSpeed Insights tool-- free, by the way, you don't need to pay someone to provide that to you-- and see how it looks. If it's bad ( and by bad I mean you see a lot of red and sub-70ish numbers), then it's probably worth spending some time fixing your major offenders.
On the other hand, if it's decent (above 80), you can still look for anything glaring that it picks up, but you're probably in pretty decent shape and might want to spend your time and money on a good content strategy, testing ads, evaluating your funnel, doing A/B tests on CTAs, etc.
The #1 candidate for quick improvement
Images. That's almost always the first problem. Be sure that you're using the right formats-- webp is usually a good choice. Then, make sure that you're serving up appropriately sized images. It is very common to see someone serving a 2000 pixel wide image in a 200 pixel wide spot. That's 100x more pixels than you need. Finally, look at using media-queries to serve up the appropriate things for different screens-- desktop vs. mobile.
Don't obsess about speed. It's a distraction. And a cop-out.
Again, recently, I received a page speed report from a client's SEO vendor with the request that I "fix" the stuff on it.
The #1 recommendation? Reduce the work on the main thread. Great. I'll get right on it. The next one? Reduce unnecessary Javascript. The #1 candidate? The Google Tag Manager code.
Not only do those recommendations not mean much, they're likely to break your site. Remove your Tag Manager Javascript, and you lose your Analytics. Remove the "unnecessary CSS" that it recommended, and the fonts all turn to Arial.
Reduce the work on the main thread? Like how? By not doing what it is we want the site to do? So that it's faster? Why not just make it a page of text? That'd be super fast. No one would want to look at it, but it'd be REALLY fast.
Focus on what really matters-- relevance, content quality, conversions. If you can make your pages faster, great! But if you can't without negatively impacting those other factors, be very sure you understand the trade-offs that you're making.