The problem with fussing over site speed

Apps that promise to improve page speed are often more harmful than we think.

They'll promise to get you a 99 page speed score. 🤮

Yet these apps and so called experts try to trick you into stressing so much about a silly score that you throw money at the problem. They don't teach you what the problem is, how to fix it, or even IF you can fix it.

They are so problematic, that Shopify's Performance Team finally cracked down on these apps.

As stated in a recent blog post, they found "15% of extensions that promised one-click optimizations" were cheating performance metrics.

My first response: ONLY 15%?

Their article explains the different ways people cheat on their performance simply to get a higher score.

Scores are just scores and don't actually translate to reality.

Page speed scores are standards trying to take a complex thing (how long it takes for the page to load in reality) and distill it down into a single number. These tools are written by humans and as such are always flawed. They are meant to guide decision making, a goal post if you will. Not a requirement you must hit in order to be successful.

The scores evolve over time too and what was perfect yesterday, might be absolutely wrong today. Because perfection is not set in stone, it changes. That's why progress towards an improved website will always outperform perfection.

Every feature you want costs you performance. Evey slider, image previewer, variant switcher, they all have performance costs.

As a rough estimate, 20% of performance issues are from Shopify which you can't change. Every platform has performance issues so don't tell me Shopify is the problem. 50% of performance issues are your theme. The remaining 30% are the bells and whistles from themes and apps. These may not always be essential to your store, but you choose to include them for various reasons.

So when you install a new theme or app, check the site performance before and after. Ask them what the impact on speed is before installing. They should be able to tell you without hesitation. Then decide if that change is worth the performance hit.

For example:

JSON-LD or SEO has no real impact on your store’s page speed or performance. With Shopify's caching off, we've found it takes 1-2% of the total page performance. All live stores run with caching on, so the actual performance impact is even lower (could even be zero).

Here we are at the end of the article and we haven't even addressed the worst part.

The problem is that this all focuses on a stupid score.

The goal of improving your site performance (not page load but overall performance) should be an improved user experience.

No one cares about your page speed score. Not your customers and not Google. They care that the website is usable and that they can find what they are looking for.

Focus on your customer and stop worrying about what Google does or does not see.