Shopify Speed Optimisation: The Subtraction Method (2026)

Published: August 24, 2026

Last updated: August 24, 2026

7 min read

You ran your store through PageSpeed Insights, and it came back red.

Deep breath.

Every Shopify store owner has stood exactly here, score glowing like a failed exam, wondering whether to buy a speed app or a new theme or a therapist.

Before you buy anything, here's the reframe this whole guide hangs on: your store isn't slow. Its passengers are.

Shopify's own platform is quick. World-class CDN, automatic image formats, hosting you couldn't improve if you wanted to (and you can't).

What's slow is everything bolted on afterwards: the app scripts, the tracking pixels, the carousel, the ghost code from things you uninstalled in 2024.

Which means the method that works is not addition.

Nobody speeds up a Shopify store by installing something.

This is the speed chapter of my complete Shopify SEO guide, full length: what to measure and what the numbers mean, the subtraction sequence, the handful of additions that survive scrutiny, and the honest ceiling nobody selling speed services will tell you about.

Short on time? Here are the key takeaways

Your store isn't slow, its passengers are: app scripts, pixels and ghost code from things you uninstalled.
Subtraction beats optimisation: nothing you install makes a Shopify store faster.
Judge field data, mobile first: the red lab score is a stress test, not a verdict.
Speed is a conversion lever wearing an SEO costume: optimise it for the till.

First, why you're doing this (it's not rankings)

Let's set the stakes accurately, because speed is the most oversold ranking factor in SEO.

Google uses page experience as a lightweight signal. A tiebreaker, not a lever.

Moving your score from 55 to 85 will not visibly move your rankings, and anyone promising otherwise is selling something.

However, speed is a heavyweight conversion factor.

Buyers abandon slow product pages, mobile shoppers most of all, and every second of load time bleeds add-to-carts.

You're optimising for the buyer holding a phone on 4G in a car park, not for Google.

Speed is a conversion lever wearing an SEO costume.

Optimise it for the till, and take the ranking tiebreaker as a bonus.

Measure like you mean it

Three templates tell the whole story: homepage, one collection, one product.

Run each through PageSpeed Insights and read two different things:

Field data (top of the report): what real visitors experienced over the last month. This is Core Web Vitals, and it's the only section Google's systems use. Targets: LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1. The definitions live at web.dev.
Lab data (the score): a simulated test, useful for diagnosis, worshipped by mistake. The red number that scared you is lab data. It's a stress test, not a verdict.

The distinction matters because low-traffic stores often show no field data at all, and app vendors sell against the scary lab score.

Judge yourself on field data where it exists, mobile first.

Use lab data to find the culprits, which is the next job.

The subtraction sequence

Here's the method, in the order that pays.

Notice that nothing gets installed until step four.

Step one: the app autopsy.

Open the lab report's waterfall and count the third-party scripts.

Match each one to an app, then ask each app the only question that matters: what job do you do?

No answer, no app.

On the stores I audit, this step alone is most of the win, and the full procurement logic (which jobs deserve an app, which categories to delete) is in my Shopify SEO apps guide, publishing alongside this piece.

Step two: strip the ghosts.

Uninstalled apps leave scripts in theme code, loading for customers years after the app died.

Search the theme for old app names and clear the remnants.

When I phased the SEO apps out of REHAUS, checking what each one actually rendered before and after removal is what kept the storefront stable.

Step three: re-measure.

Same three templates.

Most stores claw back more here than any paid fix delivers, for £0.

Step four: only now consider the additions.

If field data still fails after subtraction, the remaining culprits are almost always the theme and the images, covered next.

Subtract before you optimise.

Every hour spent tuning a store still carrying twelve dead scripts is an hour wasted.

The short list of real fixes

After subtraction, four things move the needle. None of them is an app.

Image inputs, disciplined: Shopify's CDN already serves WebP and responsive sizes. Your job is what you feed it: sharp originals around 2048px, not 8MB phone photos. Fix the templates' worst offenders first, per the image checks in my 40-point checklist.
The LCP element, usually your hero: the largest thing above the fold decides your LCP. One sharp, compressed hero image beats a five-slide carousel every time. Carousels are speed debt with a design excuse, and slide two has a click-through rate rounding to zero anyway.
Fonts, fewer: every custom font weight is a download that blocks text. Two weights of one family covers almost any store design. System fonts cover the rest at zero cost.
The theme itself, when it's the floor: a bloated theme sets a ceiling no tuning breaks. If you're theme shopping, Shopify's own free themes are the lean benchmark, and the Horizon family is the current generation built on theme blocks. If you're not shopping, don't replatform your design over a lab score. Fix the passengers first.

That's the list.

Notice what's absent: preload plugins, lazy-load wrappers, "minification" apps. JavaScript installed to remove JavaScript, all of it.

What you cannot control (stop paying people who say otherwise)

Shopify owns the hosting, the CDN, the servers and the checkout.

You cannot upgrade them, tune them or buy your way around them, and this is mostly good news: the infrastructure layer of your speed is handled by one of the best-resourced platform teams in commerce.

It also means two red flags are worth naming:

Anyone selling "server-side optimisation" for Shopify: there is no server side available to them. You're buying a story.
Chasing a 100/100 score: the last twenty points live in code Shopify controls and scripts your business actually needs (analytics, support chat, the reviews platform that earns its slot). The target is "not slow where buyers feel it", not a trophy number.

Perfection isn't available on this platform. Fast enough everywhere that matters absolutely is.

The quarterly pulse

Speed work decays the same way content does: quietly, through accumulation.

Fifteen minutes a quarter, alongside the indexation and app checks from the checklist: re-run the three templates, diff the script count against last quarter, and ask any new passenger for its job description.

Stores don't get slow in a day.

They get slow one harmless install at a time, and the quarterly pulse is what catches it while it's still one deletion away.

Final Thoughts

Speed on Shopify is a subtraction discipline sold to store owners as a shopping opportunity.

Measure the three templates. Interrogate every passenger. Strip the ghosts, feed the CDN properly, and let one sharp hero do the job of five slides.

Then bank the result where it actually pays: fewer abandoned product pages and more completed checkouts, with the ranking tiebreaker thrown in.

Speed diagnostics are part of my technical SEO audit, and the app autopsy comes standard in the free version: structure, metadata, schema and content quality, findings within five business days.

Get in touch, or email me directly.

Author Bio

Graeme Whiles is an independent SEO and AEO consultant at GWContent. He has worked with enterprise and SaaS brands, including Originality.ai, Connecteam, 6sense, Practice Better, and Peppr, growing organic traffic and AI search visibility across some of the most competitive categories in B2B. He also built Three Putt Golf Clothing from a blank domain as a live proof of concept for his methodology.

Shopify Speed Optimisation Frequently Asked Questions

Does site speed affect Shopify SEO rankings?

Barely, and only directly as a tiebreaker within Google's page experience signals. The business case is conversion: slow product pages bleed mobile buyers long before they bleed rankings. Optimise speed for revenue, and treat any ranking benefit as rounding.

Why is my Shopify speed score low even with few apps?

Usually three suspects: oversized image originals feeding the hero, a heavy theme loading fonts and scripts sitewide, and ghost code from uninstalled apps. Check the lab report's waterfall and remember the score is a stress test, and judge the store on field data, mobile first.

Do Shopify speed optimisation apps work?

Almost universally no. They add JavaScript to manage JavaScript, lifting lab scores while real visitors load more code. Every durable speed win on Shopify is subtraction (apps, ghosts, image weight, fonts) or a leaner theme. Nothing on that list installs.

What is a good speed score for a Shopify store?

Stop targeting the score. Target field data: LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1, on mobile, on your money templates. A store passing those with a lab score of 70 is in better shape than a lab-100 store failing them.