Shopify Speed Optimization: Why Your Store Is Slow and How to Fix It

If you’ve run your Shopify store through Google PageSpeed Insights and stared at a score of 28, you’re not alone. The average Shopify store scores 30 out of 100 on mobile — less than a third of the way to green.

The good news: a score of 60–80 is genuinely achievable without a developer, and that’s the range that actually matters for rankings and conversions. You don’t need a perfect 100. You need to stop losing customers to a two-second lag.

Every one-second delay in load time cuts conversions by 7%. At 3 seconds, 53% of mobile visitors leave before your page finishes loading. That’s not a technical problem — it’s a revenue problem.

What’s a good PageSpeed score for Shopify?

Here’s what those numbers actually mean for a real Shopify store:

Score	What it means
0–30	Where most stores start. App bloat + unoptimized images. Fixable.
30–50	You’ve done some work. Images are better. A few apps still firing.
50–70	Good. You’re ahead of most stores. Real-world speed feels fast.
70–90	Great. Requires deliberate optimization but no developer needed.
90–100	Possible, but usually requires developer-level work on theme code.

The benchmark most Shopify stores should target is 60–80 on mobile. Not 100. A real store with product images, a theme, and a few essential apps will never hit 100 — and that’s fine.

Why your Shopify store is slow

There are five main culprits. In most stores, it’s the first two doing most of the damage.

1. App bloat (the biggest one)

Every app you install can inject JavaScript that runs on every page load. Each front-end app costs roughly 3–8 mobile PageSpeed points. Add five apps like that and you’ve burned 15–40 points before a single customer has loaded your first product photo.

A study of 340,557 Shopify stores found the average store runs 2.5 apps — but stores with high traffic (50K+ monthly visitors) average 5.5 apps. The more you grow, the more apps accumulate, and the slower things get.

8.7% of stores are running duplicate apps in the same category without realising it — two email marketing tools, two review widgets. Each one firing on every page.

2. Unoptimized images

Your hero image, product photos, and collection banners are usually the single largest elements on each page — and they’re often uploaded straight from a camera or design tool at full resolution.

LCP (Largest Contentful Paint) measures how long your biggest visible element takes to load. For most Shopify stores, that element is an image. Oversized images are the most common reason LCP is slow.

3. Theme and page builder code

Not all Shopify themes are built the same. Themes built with drag-and-drop page builders load significantly more CSS and JavaScript than leaner themes — even for sections you’re not using. Leftover code from apps you’ve already uninstalled also accumulates in themes over time.

4. Third-party tracking scripts

Analytics, Meta Pixel, TikTok tracking, heatmaps, live chat widgets — each one adds an external request to every page load. They’re necessary, but they compound quickly.

5. Web fonts

Loading multiple font families, or large font files, adds render-blocking requests. It’s usually a smaller issue than the top four, but worth auditing if you’ve already fixed everything else.

The 5 fixes that actually move the needle

These are in order of impact. Start at the top.

Fix 1: Audit and remove redundant apps

Open your Shopify admin → Apps. Go through each one and ask: is this actively being used? Is there another app doing the same thing?

For apps you’re keeping, check if they need to be loaded on every page or only on specific pages (like a cart upsell app that only needs to fire on the cart page). Some apps let you control which pages they load on — check their settings.

Removing one heavy app can recover 5–10 PageSpeed points overnight.

Fix 2: Compress and resize your images

The target: no product image larger than 200KB. Hero images can go up to 500KB if they’re genuinely full-width.

Before uploading, export images in WebP format — Shopify accepts it natively and WebP files are typically 25–35% smaller than JPEGs at the same visual quality. Most design and photo tools support WebP export.

For stores with large catalogues, doing this manually isn’t realistic. Plug In Speed compresses and converts images automatically across your entire product catalogue — you set it once and it runs in the background.

Fix 3: Switch to a speed-first theme

If you’re on a heavily customised third-party theme or a page builder theme, consider migrating to a Dawn-based theme (Shopify’s native free themes are fast by default). Shopify’s own themes are built to the platform’s performance standards in a way most third-party themes aren’t.

This is the most disruptive fix — it means rebuilding your store’s look — but it delivers the biggest single improvement for stores with theme bloat.

Fix 4: Defer non-critical scripts

In your Shopify theme code (or through a speed app), you can tell certain scripts to load after the page is visible to the user — this is called deferring. Non-critical scripts like heatmaps, chat widgets, and review widgets are good candidates.

If you’re not comfortable in theme code, a speed app handles this automatically.

Fix 5: Enable lazy loading for images

Lazy loading means images only load when a visitor scrolls to them — instead of loading every image on the page the moment someone lands. Shopify 2.0 themes have basic lazy loading built in, but it doesn’t cover all image types (collection grid images, slideshow banners).

So even if you have 40 products on a collection page, the above-the-fold content loads first and fast — the rest follows as the visitor scrolls.

How quickly will you see improvements?

Merchants who fix their top issues — usually app bloat plus image compression — typically see PageSpeed score improvements within 24–48 hours (scores update based on real-world data). Improvements in Google rankings and conversion rates take 2–4 weeks to show up in analytics, since both respond to sustained performance data, not a single measurement.