● Findings · October 2026 · Daniel the Tuner

Shopify speed statistics 2026: what 5,991 live stores tell us

How fast is a typical Shopify store, really? This month we ran Google PageSpeed on the homepages of 5,991 live stores, most of them drawn at random from the world’s top million websites. Here’s what came back.

The typical store scores 56 on mobile

Across all 5,991 stores, the middle score on Google’s phone test was 56. On desktop it was 76.

• 35% scored under 50 on mobile.
• 63% landed between 50 and 89.
• 1.7% reached 90 or more. On desktop, 26% did.

So if your store sits in the 50s on mobile, you’re in very ordinary company. A 90+ mobile score is rare on a real store with real products and apps.

Bigger brands score lower

We split the stores by where they came from. The pattern is consistent: the more established and busy the store, the lower the score.

Typical (median) Google PageSpeed mobile score. Bar length is the score out of 100.

Established brands run around ten apps on the homepage, and the phone spends almost a full second busy with scripts before the page responds (914 ms of blocking time, against 86 ms for newer stores). Among the 134 stores in our sample that rank in the world’s top 100,000 websites, the typical score was 50, and half scored under 50.

More apps, lower scores

The clearest pattern in the data: the more third-party apps on the homepage, the lower the score, and the steeper it gets.

Typical mobile score and blocking time by number of apps found on the homepage (analytics tags not counted). 5,895 stores with a standard theme.

This is a link, not proof that each app costs a fixed amount. Stores with more apps also tend to be bigger, with heavier pages. But the phone’s busy time grows about eightfold from the first row to the last, and that is time a shopper spends tapping a page that doesn’t answer yet.

Apps aren’t the enemy here. Many of them pay for themselves. The useful question is which ones are worth their weight on the homepage, and which could load later or live somewhere else.

What keeps the phone busy

We measured how long each third-party tool keeps a phone’s processor busy during Google’s test, for every tool that showed up on 25 or more stores. Here they are grouped by type, as the typical cost per tool.

Typical main-thread time per tool on Google’s mobile test. Ranges cover the different tools within each type.

Two things stand out. Fraud screening and spam checks cost around a full second each, and they start working as the page loads, long before anyone reaches a form or the checkout. And the tag manager is everywhere: it showed up on 43% of stores, at around 280 ms each.

A low score isn’t always a slow store

Google publishes two kinds of numbers. The PageSpeed score comes from one simulated visit in Google’s phone test. Separately, Google collects how fast pages actually load for real Chrome users over 28 days, and checks three things: how fast the main content shows, how fast the page responds to a tap, and whether the layout jumps around.

4,998 stores in our sample had enough visitors for Google to report real-user data. For them, the main content took a typical 9.9 seconds to appear in the lab test, but 1.9 seconds for most real shoppers.

So before you spend money chasing a number, look at the real-user section of your PageSpeed report (it’s at the top, when your store has enough traffic). If it passes, speed work still helps shoppers on older phones, but it’s rarely an emergency.

Code that outstays its app

• 13% of stores appear to request a script that no longer exists, and for 6% we confirmed the file is gone. The browser still waits for the answer.
• 57% of stores have app code written straight into the theme files, where uninstalling the app never reaches it. Among established brands and the busiest sites, it’s 83%.

We wrote up how to find and safely remove it: How to find leftover app code on Shopify.

How do popular themes compare?

Nine themes had more than 100 stores each in our sample. Most landed within a few points of each other.

Typical Google PageSpeed mobile score by theme, themes with more than 100 stores. Shown in one color on purpose: this is a comparison of stores, not a ranking of themes.

A gap of a point or two is within the normal run-to-run wobble. These are real stores, each with its own apps, images and content, so this isn’t a test of the themes on identical pages. We can’t tell from this data how much of a gap comes from the theme and how much from the kind of store that picks it. What the data does show is that the apps and content added on top move the score far more than the choice between these themes.

What to take from this

1. Check your real-user data first. It tells you whether shoppers feel a problem, which the score alone can’t.
2. Count what loads on your homepage. Each app there should be earning its place. Some can load later, or only on the pages that need them.
3. Look for the expensive surprises: a fraud or spam check on the homepage, an embedded video, a heavy pop-up, a tag manager carrying more than it needs to.
4. Clear out leftovers from apps you no longer use.

How we got these numbers

In October 2026 we measured more than 6,200 live Shopify stores in two batches, and 5,991 of them returned a mobile score. The larger batch is a random sample of the Tranco list of the world’s top million websites (list N26XW): we checked which of 300,000 randomly chosen domains are hosted on Shopify, then measured 4,840 of them. The earlier batch came from three public sources: a list of 265 well-known direct-to-consumer brands (112 of them on a standard Shopify theme), live stores featured in the Shopify Theme Store, and stores their owners posted in the Shopify Community for feedback. Each store got one Google PageSpeed Insights run on mobile and one on desktop. Stores with no detectable Shopify theme are counted as headless. Apps were identified from the page code, the same way our free check does it, with analytics tags left out of the app count. Third-party costs use PageSpeed’s per-script main-thread time and include only tools seen on 25 or more stores. Real-user figures come from Google’s Chrome UX Report as returned by PageSpeed Insights. A single PageSpeed run can move by several points, so these numbers describe patterns across many stores, not any one store. We don’t name any of the stores or apps.