How to Choose the Right Shopify Apps Without Slowing Down Your Store

April 10, 2026
App Guides
10 min read
a speedometer gauge with app blocks affecting the speed needle

Stores with 14+ apps installed often see product pages taking 6+ seconds to load on mobile. The fix is not to remove all apps. It is to remove the ones built wrong. Choosing Shopify apps without slowing down your store comes down to understanding how apps inject code, when they phone home to external servers, and what the actual cost is per request.

Most merchants install apps the way people add browser extensions. One for reviews, one for upsells, one for swatches, one for pop-ups. Before long the storefront is loading 9 extra JavaScript files, 4 external stylesheets, and making 11 third-party API calls on every single page load. The customer doesn’t see any of that. They just see a slow store and leave.

You can run a store with 10+ apps and still hit sub-2-second load times. But only if you pick the right ones. Here is how to evaluate apps before you install them, and how to audit what you already have.

In this post

Why apps slow down Shopify stores

Shopify’s Liquid rendering is server-side. Fast by default. The problem starts when apps inject client-side code that runs after the page has already loaded (or worse, blocks rendering until it finishes). Each app can add JavaScript, CSS, fonts, tracking pixels, and external API calls. Individually these might add 50-200ms. Stack six of them together and you are looking at an extra full second, sometimes more.

Google’s Core Web Vitals penalize this directly. Largest Contentful Paint, Interaction to Next Paint, Cumulative Layout Shift. Apps that load heavy scripts or rewrite DOM elements after render will hit all three. And Google doesn’t care that you “need” the app. It just scores the page.

The tricky part? Shopify doesn’t show you a per-app performance breakdown in the admin. You have to dig for it yourself. Use our free App Detector to see what apps a store is running, then cross-reference with Chrome DevTools or WebPageTest. That alone will tell you more than any app listing page ever will.

Three app architecture patterns and their speed cost

Not all apps are built the same way. Understanding these three patterns will instantly make you better at spotting fast vs slow apps.

Pattern 1: Metafield-based (fastest). The app writes data to Shopify metafields during setup. On the storefront, a small script reads those metafields and renders. No external API calls. No waiting on a third-party server. The data loads with the page itself. Apps like Rubik Variant Images use this approach, which is why they don’t add measurable load time to product pages.

Pattern 2: API-call on page load (slow). Every time a customer visits a page, the app’s JavaScript calls an external server to fetch data. Reviews, recommendations, pop-up rules, A/B test configs. If that server is slow (or down), your page waits. If the customer is in Australia and the app server is in Virginia, add 300ms of latency. Minimum.

Pattern 3: Iframe or third-party widget (slowest). The app loads an entire widget from an external domain. Chat widgets, social proof pop-ups, embedded forms. These pull in their own CSS, JS, fonts, and sometimes even their own analytics. They are basically tiny websites loading inside your website. Why does anyone still build apps this way? I genuinely don’t know.

Red flags to check before you install any app

Before hitting “Add app,” look for these signals. They won’t guarantee the app is slow, but they should make you pause and investigate.

  1. No mention of page speed anywhere. If the app listing doesn’t mention performance at all, the developer probably hasn’t thought about it. That is a problem.
  2. Reviews mentioning “slowed down my store.” Filter by 1-2 star reviews. Speed complaints are the most reliable signal you will find.
  3. The app injects a visible widget on every page. Pop-ups, notification bars, chat bubbles. If it renders on every page, its code loads on every page.
  4. “Powered by [AppName]” badge. That badge loads from an external domain. It is tracking you, and it is slowing the page.
  5. Shopify plan-based pricing. This doesn’t directly correlate with speed, but it signals a less technical team. Flat-rate apps tend to come from developers who invest more in architecture.

Run your store through the Store Analyzer to get a quick snapshot of third-party requests your site is making. That data is worth more than any app listing screenshot.

How to audit your current app stack

Already have apps installed? Good. Here’s the process I use when merchants ask me to clean up their stack. Takes about 30 minutes.

Step 1: List every installed app. Go to Settings > Apps and sales channels. Write them all down. Include apps you installed months ago and forgot about (you probably have at least one).

Step 2: Check for leftover code. Uninstalled apps sometimes leave theme code behind. Open your theme code editor and search for script tags or Liquid snippets that reference app names you no longer use. I once found code from three different review apps in a single theme. The merchant was only using one of them. The other two were still loading JavaScript on every page.

Step 3: Run a waterfall test. Open your product page in Chrome DevTools (Network tab) and reload. Sort by size or time. Look for requests going to domains you don’t recognize. Each external domain is likely an app phoning home. Count them. If you see more than 5 external domains loading on your product page, you have a problem worth solving.

Step 4: Disable one app at a time. Measure page speed with Google PageSpeed Insights before and after. Tedious? Yes. But it shows you exactly which app costs the most. Some stores find that a single abandoned-cart pop-up app adds 1.8 seconds to every page load. Removing it and their SEO scores jumped 15 points overnight.

For a deeper audit framework, check our app stack audit guide. It covers the full methodology with real examples.

App categories that hurt performance most

Some app categories are worse than others. Here is the rough hierarchy based on what I have seen across hundreds of store audits:

Worst offenders: Live chat widgets, social proof pop-ups (“John from Texas just bought…”), and page builder apps that load their own rendering engine. These almost always use Pattern 3 (iframe/widget) and they run on every single page.

Moderate impact: Review apps, recommendation engines, and A/B testing tools. They make API calls per page load but most have gotten better about lazy loading. Still, they add up fast if you stack them.

Low impact (when built right): Variant image apps, swatch apps, SEO tools, and bulk operations apps that only run in the admin. If the storefront component uses metafields instead of API calls, these barely register on a speed test.

The lesson? Not every app category deserves the same scrutiny. Focus your audit energy on the worst offenders first.

What fast apps look like

Fast apps share a few traits. Look for these when evaluating:

They store data in Shopify metafields or metaobjects, not on external servers. They render using small, scoped JavaScript (or even pure Liquid). They don’t load fonts, stylesheets, or tracking scripts from third-party CDNs. They scope their CSS so it doesn’t conflict with (or override) your theme. Shadow DOM is the gold standard here.

They only load code on the pages where they are needed. A variant image app that loads its script on collection pages where it does nothing? That is wasted bytes. A well-built app loads nothing on pages where it has no work to do.

And honestly, you can tell a lot from how the app’s admin UI feels. If the app’s own dashboard is sluggish and bloated, the storefront code probably isn’t any better. Fast teams build fast products. Slow teams build slow products. It sounds reductive but it holds up.

A practical app selection checklist

Use this before installing any new app. Print it out, bookmark it, tattoo it on your arm. Whatever works.

  • Does the app use metafield-based rendering or external API calls?
  • Does it load scripts on every page or only where needed?
  • Do the 1-star reviews mention speed issues?
  • Does the app have “Built for Shopify” certification?
  • Is pricing flat-rate or plan-based? (preference, not dealbreaker)
  • Can you test it on a dev store before installing on production?
  • Does the app’s uninstall process remove all injected code?

Two more things worth checking: run your product page optimization checklist before and after install. And test on mobile, not just desktop. Most Shopify traffic is mobile, and slow JavaScript hits mobile devices much harder than desktop.

If you run stores with many product variants, look into how combined listings apps handle rendering. The good ones (metafield-based, shadow DOM isolated) add virtually nothing to page load. The bad ones make an API call for every swatch on every collection page card. That difference can mean hundreds of extra requests on a collection page with 48 products.

FAQ

How many Shopify apps is too many?

There is no universal number. I have seen stores with 20 apps loading in under 2 seconds, and stores with 5 apps that take 7 seconds. It depends entirely on how each app is built. Focus on architecture quality, not app count.

Do apps still slow my store after I uninstall them?

Sometimes, yes. Some apps inject code into your theme files and don’t remove it on uninstall. Check your theme code for leftover script tags and Liquid snippets after removing any app.

What does “Built for Shopify” mean for performance?

The Built for Shopify badge means the app passed Shopify’s quality and performance review. It is not a guarantee of zero speed impact, but it filters out the worst offenders. Prefer certified apps when possible.

Should I avoid all apps that make external API calls?

Not necessarily. Some app categories (like reviews) genuinely need external data. The question is whether the app lazy-loads that data or blocks page render while waiting for it. A well-built review app loads reviews after the main content is visible. A poorly built one blocks the entire page.

How do I measure the speed impact of a specific app?

Run Google PageSpeed Insights on a product page. Note the score. Temporarily disable the app (or remove its theme code). Run the test again. The difference is that app’s cost. Do this for each app you suspect and you will have a clear picture within an hour.

Want to see how a metafield-based app handles variant images without touching your page speed? Try Rubik Variant Images free on the Shopify App Store.

Co-Founder at Craftshift

You may also like