No, You Will Never Pass Google’s Mobile Core Web Vitals Assessment

Unless you manually hack together your mobile site out of bare HTML and show no pictures.

When I started out, I was going to write an interesting article about how you could get a website to pass the mobile Core Web Vitals assessment that Google is going to roll out as a major driver of SEO ranking in the coming months:

Isn’t is already May 2021?

So, hey, we developers better look lively and get our mobile Core Web Vitals in shape. I mean, how hard can it be?

I’m a big fan of using Vue.js with Nuxt.js. I think the whole point of that is being both effective and quick, am I right? I created a speedy JAMStack app, generated static HTML, and used what I think we can all agree is a cheap but fast server to host it:

Express. Sounds fast

And the results were pretty good:

92 seems a bit disappointing though.

Not bad, right? We are in the zone. Although, given that we are just showing an inline SVG, two buttons, and a bit of text, you would have hoped for more like 99? Even 100? Seems we could have reduced the load time by removing some of the unused JavaScript. Well, we are pretty much stuck with that script because without it, no framework and no developer fun happening at all.

So far so meh. But suppose you wanted to, oh I don’t know, use a beautiful Material UI design framework to help you produce a site people might actually want to visit? Perhaps we might include Vuetify in the build — then what happens?

Oh. Oh I see

This is not what we came here for at all. Remember, this is a website running locally, in optimum conditions, running an app optimized for the Jamstack, the most scorchingly fast way of serving a site there is. This means that, before we have even written a line of code, before we have presented all our client’s beautiful media and well-chosen words, we have already failed, and our SEO is down the tubes. This, good people, is wrong.

Maybe if we ditch Vuetify and we just go for something more trendy and less weighty we can solve our problems? How about Tailwind CSS?

Well, the picture is better, but we are already out of the top echelon, under 90, and again, we have not even started coding at this point.

Which got me thinking, how do the real pros do it? They must be such awesome coders to have solved this seemingly intractable problem. How, for example, does the home page of Medium stack up?

Ouch

35. Wow, that seems like not even a medium score.

So, the folks at Google must know what they are doing, right? They make a lot of money. Let’s see how they will judge one of their own sites. YouTube is just about the most popular mobile site on the whole Internet, visited by millions every day.

Will you look at that First Contentful Paint timing?

I wonder how much it hurts the devs at Google to be scolded by their own Page Speed Insights about the things they could do to improve this. I wonder how sad they are when they peruse their Google Search Console and are greeted with the sort of ill-news most websites must be facing:

Oh my poor URLs

And all because your Cumulative Layout Shift was not what it could have been.

In case you think there’s just some problem with Lighthouse on my laptop, here is Page Speed Insights’ assessment on the same page, which is even more dismal.

We have not even started looking at websites that are actually slow. You can visit thousands of beautifully put-together sites that are the most popular on the web, seem super-fast, and even enjoyable to use on your phone, and yet nearly all fail the Core Web Vitals mobile test, and fail it miserably (the desktop version is much easier to do well on).

Now, none of this would be a problem, but SEO guys all over the world are hearing that Core Web Vitals are going to start hitting their site rankings any day. They are, as I write, rushing to the developers and screaming “Something’s gotta be done about it!” And, in many cases, it will not go down well when their developers shrug their shoulders and admit that nothing can be done.

I don’t have any answers, I’m afraid. You probably can’t make your website pass this test unless you are prepared to compromise drastically on the architecture and the content.

I don’t understand why Google has announced that mobile Core Web Vitals are going to start affecting SEO when the system seems so broken. I don’t understand why the scoring for mobile Core Web Vitals is set up to make a reasonable score so unattainable.

All I know is that, as an independent developer, it is has already cost me a lot of time and money to deal with the fallout, and doubtless will for some time to come.