How to Speed Up a Slow Shopify Store: Measure, Then Fix

The Themes page in your admin shows Moderate or Poor next to loading speed. On a phone, product pages hang blank and then appear all at once. Or the store felt fine until the week you installed an app or edited the theme.

Each of these has a cause you can find and a fix you can undo, as long as you measure before you change anything.

This guide follows that order. It shows where Shopify measures your storefront and how to read the result, what slows Shopify stores most and which fixes to try first, when a new theme or a developer is worth it, and how to make each change one at a time with a theme backup saved, so that a fix that doesn't help can be undone.

Where to Measure Your Store's Speed

Shopify gives you one measuring tool inside the admin, and Google gives you two outside it. They answer different questions, so it helps to know which number you are looking at before you act on it.

Open the Web performance report

Shopify's own measurement lives in the Web performance reports. You can open the Over Time reports from the performance metric summary on the Online Store > Themes page, and every web performance report from Analytics > Reports, where typing a colon (:) into the search shows them all. Staff need the Reports permission to open them.

The summary on the Themes page rates three Core Web Vitals (loading speed, interactivity and visual stability) as Good, Moderate or Poor, based on real visitors over the past 30 days, and a date picker lets you switch it to Today, Last 7 days or Last 30 days.

Two Shopify pages describe the window differently: the reports page gives the summary those 30 days, while the web performance overview describes a performance score that combines your home page, most visited product page and most visited collection page from the last 7 days (both read October 15, 2026). Go by the window printed on your own report rather than either page.

Once the summary has enough data to show a score, the reports narrow the problem down. There are three kinds:

In Shopify adminOpen in your store

1. Analytics
2. Reports

Zoom

Three limits shape how you use the reports. The data might be delayed by up to 36 hours, so a change made this morning won't show today. Only the last 90 days are kept, so note your baseline before it rolls off. And the report needs real visitors: a store in private mode, or one with no page views yet, has no data at all, while a store with little traffic sees larger fluctuations.

What Good, Moderate and Poor mean

The rating reflects the top 75% of your visitors' experiences, so one fast test on the office connection tells you little. Shopify's overview adds that for your performance to count as good with search engines such as Google, at least 75% of your page loads need Good scores across all three Core Web Vitals. Shopify's developer documentation says Google uses these metrics as a ranking signal, and its help center says pages that perform poorly across the Core Web Vitals might be ranked lower.

Mobile is the harder test. Shopify's theme documentation says Core Web Vitals scores are measured on mobile by default, and that mobile devices account for the majority of Shopify storefront traffic, on slower CPUs, constrained memory and variable network connections. Read the mobile figures first.

Why PageSpeed Insights shows different numbers

Google's PageSpeed Insights shows two things on one screen. The top block is field data: real users' First Contentful Paint, Interaction to Next Paint, Largest Contentful Paint and Cumulative Layout Shift over the previous 28-day collection period, taken from the Chrome UX Report (CrUX). Below it sits a lab test, a simulated load of the page on a single device and a fixed set of network conditions.

• Web performance report (admin): spot a problem among your own visitors, and confirm a fix worked.
• PageSpeed Insights, field block: a second view of real Chrome users over 28 days.
• PageSpeed Insights lab test, or Lighthouse in Chrome DevTools: debug one page right after a change, without waiting for visitors.
• Chrome DevTools Performance panel: confirm the root cause the report points to.

A few more differences explain most gaps. A green lab score of 90 or more is considered good, but Google notes that good lab data does not necessarily mean real users have a good experience. Shopify's developer documentation warns that Lighthouse scores can differ significantly from real-user data, that CrUX aggregates to the 75th percentile and lags by weeks, and that CrUX needs sufficient traffic to populate. It calls the disagreement normal, usually a difference in what is being measured rather than a broken tool.

Shopify's own pages disagree about where its report's data comes from. The help center's overview says the reports gather a broader data set from all Chromium browsers (Chrome, Opera, Edge, Samsung Internet) and Firefox, while CrUX covers opted-in Chrome browsers only. Shopify's developer page on testing for performance says the data is collected from Chrome users who have opted into sharing performance metrics (both read October 15, 2026).

Either way, treat the admin report as the record of your store and PageSpeed Insights as a second opinion, and don't try to reconcile the two figure by figure.

Is Your Store Slow Enough to Fix?

Shopify's overview names four situations where improvement deserves your time:

• Your scores are consistently in the Poor range across multiple metrics.
• You notice a significant drop in performance without making changes to your store.
• Customer feedback points to slow loading times or a poor experience.
• Your conversion rates are declining alongside your performance scores.

Outside those, Shopify's advice is calmer: as long as the overall metric is good, don't worry too much if some dates or pages fall into the Moderate or Poor categories. When a ranking is Moderate or Poor, the reports are where you find out why.

If the fourth situation is the one that worries you, what faster pages are worth in conversions is covered in our guide to how the best Shopify stores are built.

None of the four mentions a target score. Shopify's overview says there is no such thing as a perfect speed score, and that the closer you get to 100, the more difficult it becomes to improve further. Its advice is direct:

No help center page we read sets a PageSpeed or Lighthouse score for a merchant's store to reach (see the search boundary further down). The targets Shopify does publish are the Core Web Vitals limits above and the entry requirements for themes and apps.

Find the Cause Before You Change Anything

The cheapest fix is the one aimed at the real cause. Shopify's performance team puts the order of tools in one line:

Find the row that matches what your report shows:

In Shopify adminOpen in your store

1. Analytics
2. Reports
3. Largest Contentful Paint (LCP): Over Time

Zoom

One method in this guide is ours rather than Shopify's: switching off a suspect app's embed and comparing the page before and after. No help center or shopify.dev page we read describes that as an official way to find a slow app. Shopify applies the same before-and-after idea to apps that affect the storefront directly, measuring Lighthouse scores before and after an app is installed, and the safe way to run it on your own store is the one-change process at the end of this guide.

What Slows a Shopify Store Most, and What to Fix First

Shopify's help center lists five areas that influence storefront speed: installed apps, third-party libraries and services, analytics libraries, your theme's code, and the number and size of images and videos. Its page on improving web performance narrows that to the three biggest factors for Shopify stores: the theme, installed apps, and third-party code, including tag managers.

Shopify lists the factors but doesn't number a repair order, so the order below is ours. It starts with what is cheapest to test and easiest to undo, and with what Shopify's theme documentation calls the highest-impact optimization for responsiveness: auditing and removing third-party scripts. Replacing the theme comes last, in its own section, because it is the most expensive move.

If your row in the table above points at a cause, start there; use this order when the report doesn't single one out. Either way, duplicate the theme and record a baseline before you touch the live store.

Shopify Academy's short video from 2022 walks the same ground: limiting third-party apps, optimizing theme code, getting theme support and managing media. It talks about the single store speed score, which Shopify has since replaced with three Core Web Vitals scores, so take its advice and set aside its references to the old score.

Apps, pixels and tag managers

Shopify's advice is to evaluate every installed app and every piece of third-party code and ask whether it creates enough value to offset the speed it may cost. Too many apps loading assets on your storefront can slow performance, and tracking works the same way: Shopify's pixels documentation says plainly that the more scripts you have, the slower your store will run.

Tag managers deserve their own pass, because Shopify recommends auditing yours to remove any unused or low-value tags. If you run Google Tag Manager, our Google Tag Manager guide for Shopify covers where its tags load and how to set it up.

Many apps plug into your theme as app embeds, which you can activate and deactivate in the theme editor. That matters for speed: Shopify tells app developers to use app embed blocks because they can load scripts only on specific pages. Some apps still inject code directly into your theme's code; Shopify's help describes the activate and deactivate toggle in the theme editor for app embeds, not for that injected code.

An app's place in the App Store tells you less about your store than it seems. Shopify's best practices say an app shouldn't reduce Lighthouse performance scores by more than 10 points, and for apps that affect the storefront directly Shopify measures that before and after installing the app; its guide for developers running a similar test says to start from a clean install of a supported Shopify theme, such as Horizon, without their app or any other apps installed. Shopify publishes that limit as a best practice for App Store apps and makes it a requirement for the Built for Shopify badge.

Your store, with its own theme and its other apps, is a separate test, and your Web performance report measures it with your real visitors.

What the Built for Shopify speed criteria do and don't cover is set out in our guide to the badge.

Page builder apps are apps too, and whether a page builder will slow your store has its own section in our landing pages guide.

If a page still hangs as a blank page and then loads all at once, Shopify suggests using your browser's developer tools to identify slow-loading scripts and contacting the app or theme developer. When the problem is coming from an app, its developer is the person to ask.

Code that uninstalled apps leave behind

Removing an app in the admin doesn't always finish the job. Some apps add code to your online store theme that isn't automatically removed when you uninstall the app, and that leftover can go on loading on your pages. Shopify requires developers whose apps leave code behind to give merchants instructions for removing it completely, so asking for them is reasonable.

1. Open a page of your store, view the page source in your browser and search for the removed app's domain. If it still appears, the app's scripts are still loading.
2. Open App embeds in the theme editor and check that nothing belonging to the removed app is still switched on.
3. Ask the app's developer for removal instructions, and back up the theme before you follow them.

If the store has been through several developers or many apps, tracking down the ghost code apps leave is covered step by step in our store code audit guide.

Trim the theme you already have

Much of a theme's weight is set in the theme editor rather than in code. Shopify's help center says high numbers of sections on pages can lower web performance. Shopify's guidance points to four moves:

• Fewer sections per template. Avoid adding too many sections to the page templates in your theme.
• Pagination on large collections. If a collection holds a large number of products, consider turning on pagination so fewer products load at one time.
• The latest theme version. Shopify's performance team tells merchants to stay on the latest version of their theme to get its optimizations; whether a Shopify theme updates itself depends on where it came from.
• An audit of customizations. Shopify's performance team lists auditing your theme's customizations among the next steps for merchants, alongside checking apps and tags.

In Shopify adminOpen in your store

1. Online Store
2. Themes
3. Edit theme

Zoom

Images: what Shopify handles and what you control

Shopify already does much of the image work for you, so check what it handles before you redo any of it by hand. The split looks like this:

The 200 KB figure and the advice against manual compression are not in conflict. Shopify's theme images page says images don't need to be manually compressed before uploading, while its product photography guide says a good image size to aim for is no larger than 200 KB, with lower numbers being better. One is about running files through a compressor, the other about the size you upload: a photo much wider than the slot it fills still carries weight the visitor never sees.

The one image that must never be lazy-loaded is the main image on the first screen, usually the hero banner or the first product photo. Shopify's theme documentation says to load images in the initial viewport eagerly, and notes that lazy-loading libraries such as lazysizes hide the image URL from the browser's preload scanner until JavaScript runs.

In HTTP Archive data for Shopify pages on mobile, published by Shopify's performance team in August 2023, Shopify sites that lazy-loaded their LCP image had a median LCP 1.0 second slower than sites that loaded it eagerly. That is a correlation across sites, not a controlled test, but it points the same way as the documentation.

Format support has its own answer: which image formats Shopify accepts, including where AVIF fits.

For hero banners in particular, what size a Shopify banner image should be depends on how the theme section crops it.

Should You Switch Themes to Get Faster?

A new theme is the most expensive speed fix, so it comes last. Shopify's position is that all the Horizon family of themes, and its Online Store 2.0 themes, are free and optimized for web performance, and that at scale a well-optimized base theme is the biggest factor in final store performance. No Shopify page we read says Horizon is faster or slower than Dawn overall; a September 2026 post from Shopify's performance team compares how much each theme gained from streamed HTML, not their overall speed.

On the Shopify pages we read, the closest thing to an overall comparison is field data. Shopify's theme performance table showed, as of October 13, 2026, that 92.2% of tracked Dawn stores and 88.3% of tracked Horizon stores passed all three Core Web Vitals, counting a store as passing when at least 75% of its visitors had a good experience. The table counts stores, not scores; it mixes all devices, and it includes only stores with at least 100 views a day, a custom domain and one of the latest three versions of the theme. Read it as a sign that both themes work well for most stores, not as a ranking.

How Horizon and Dawn compare on speed, including independent demo-store tests, is weighed in our Horizon guide, and its verdict matches this one: moving to Horizon is not a guaranteed speed win.

If you do switch, plan for two side effects:

• Apps don't follow you. Shopify says apps you installed for your theme won't be active by default when you change your published theme, so each has to be re-activated in the new one.
• Your old theme becomes the way back. When you publish a new theme, the previously published one moves to the Draft themes section, where you can publish it again.

What Shopify Already Does, and What Your Plan Doesn't Change

A good part of what generic speed checklists tell you to do is already done. Shopify's platform performance documentation lists the optimizations its infrastructure applies to every storefront, and its help center adds a few more:

This is also why third-party speed scanners can mislead. Shopify notes that many of the recommendations such tools make are already implemented by Shopify or don't apply to Shopify stores, so check a scanner's to-do list against the table above before anyone bills you for it.

Your plan is the other thing people expect to matter. No Shopify page we read ties storefront speed to the plan or limits the Web performance report by plan (search boundary below). What the plan does change is capacity: Admin API rate limits for apps vary by plan, and Shopify lists high-volume checkout for flash sales and peak traffic as Plus-only. On the Storefront API, Shopify says, buyer traffic isn't rate-limited at all.

If you are weighing an upgrade for other reasons, our plan comparison covers what each tier adds.

On Shopify Plus, checkout adds one task of its own: Shopify recommends identifying slow checkout UI extensions, removing unused ones and working with their developers to reduce checkout LCP and INP.

Some limits sit outside the platform altogether. Shopify reminds merchants that a customer's device, network and location are beyond their control, which is one more reason to judge the 75th percentile of many visits rather than a single test.

The most radical way past a theme's limits is a custom headless storefront, which swaps the theme for a development project; when headless makes sense is covered in our custom development guide.

When to Bring In a Developer, and What to Ask For

Some fixes need code: a slow script inside the theme, an image loaded the wrong way, leftover code that an app's instructions don't cover. Who to ask depends on where the problem sits:

A developer works faster from a brief than from "make it faster". Shopify's theme performance best practices give the list to start from:

• Audit and remove third-party scripts, checking coverage on each template, because a script that looks unused on the home page might do real work on a product page.
• Never lazy-load the main (LCP) image, and give it fetchpriority="high".
• Don't hide the main image behind an animation.
• Render content in Liquid and HTML rather than building it with JavaScript.
• Avoid deeply nested Liquid loops.
• Load first-paint resources before content_for_header, and use preload sparingly.

How to write a project brief for a developer, and what to check before you hire one, is covered in our guide to hiring a Shopify developer.

Make Speed Changes Safely: Measure, Change One Thing, Measure Again

Speed work goes wrong in two ways: several changes at once, so nobody knows which one helped, and no way back when a change breaks something. Shopify's developer documentation describes the cycle that avoids the first:

Real-user data is slow by design, so plan the waiting. The Web performance report can trail by up to 36 hours, and Shopify warns that, due to the way they are calculated, metrics might not change immediately after you change your store; the summary covers up to the past 30 days of visits. PageSpeed Insights field data covers 28 days, and CrUX lags by weeks.

The lab test is the only instant check, and it should run on the published store: Shopify notes that preview themes, the preview bar and the theme editor aren't streamed.

For a theme change, the way back is a copy of your theme. Shopify's help center recommends making a backup copy before you customize a theme, so you can discard your changes and start again; the copy is named Copy of and the theme's name. If your admin already holds 20 themes, you need to delete one first. Publishing the backup later moves the changed theme to Draft, where it stays available.

In Shopify adminOpen in your store

1. Online Store
2. Themes

Zoom

Work through the steps below for every change, starting at step 1 each time, so the baseline and the backup always show the store just before the change you are testing; step 7 decides whether that change stays.

The Bottom Line

Shopify names the theme, installed apps and third-party code as the biggest factors in a store's speed, and all three are in your hands. As of October 2026, no Shopify page we read ties storefront speed to the plan.

Frequently Asked Questions