A familiar complaint: the store takes several seconds to load and people abandon checkout halfway through. The catalog isn’t huge, hosting is decent, but the product page fires hundreds of database queries and the plugin list keeps growing.
This is a pretty typical picture. Nobody installs all those plugins at once – they pile up over a year or two: a shipping calculator here, a subscription popup there, a “customers also bought” widget somewhere else. Each one seems light on its own. Together they turn WooCommerce into something that loads slower than the checkout page it’s supposed to lead to.
Figure out where it’s actually slow first
Before switching anything off, you need to see the whole picture. Skip this step and optimization becomes guesswork: you disable a plugin, it feels faster, you turn it back on, and now you have no idea whether the plugin was ever the problem.
The main tool here is Query Monitor. Free plugin, takes five minutes to install, and it immediately gives you a panel at the bottom of the admin screen: how many database queries built the page, which ones are slow, which plugin or hook triggered each one, how much time went to PHP versus rendering. A normal WordPress product page without bloat runs a modest number of queries. When you count them in hundreds, that’s not a quirk of WooCommerce’s architecture – some specific plugin is doing something wasteful.
Query Monitor tags every slow query with the name of the plugin that fired it. No guessing involved, just reading.
The waterfall in DevTools
The second thing worth opening is the Network tab in Chrome DevTools, Waterfall view. This shows the frontend side rather than the backend: how many JS files load, how much CSS, whether anything is render-blocking, and which third-party domains get pulled in on the side – trackers, chat widgets, fonts from external CDNs.
Sometimes the waterfall shows a dozen separate requests to Google Fonts and the Facebook Pixel loading synchronously, before any content appears on screen. Deferring those usually cuts First Contentful Paint noticeably.
The elimination method: turn plugins off one at a time
The simplest way to find the heavy plugin is the old elimination method. Run GTmetrix or PageSpeed Insights, record the current numbers (don’t trust your memory – an hour later you’ll forget which score belonged to which state), then disable one plugin, measure again, repeat.
Do this on a staging copy, or at least outside business hours – disabling a plugin on a live store means part of the functionality, shipping calculations or payment options, temporarily disappears for real shoppers.
One more thing: clear the page cache and browser cache before every measurement, otherwise you’re not comparing plugins, you’re comparing a cached page to an uncached one.
With a long plugin list, there’s no point disabling them one by one for hours – split it in half instead. Turn off half, measure, and if there’s a meaningful improvement, keep narrowing that half down with binary search. Twenty plugins get checked in five or six rounds that way, not twenty.
The usual suspects
A handful of plugin categories show up as the culprit more often than others – not because they’re badly built as such, but because people install them and forget they exist.
- Heavy page builders (Elementor, Divi and similar) with dozens of modules enabled, even on pages that never use most of those modules.
- Recommendation and “related products” plugins that build database queries on the fly for every page load instead of caching the result.
- Analytics and tracking piled up in layers – Google Analytics, Facebook Pixel, Hotjar, a couple of ad tags and an email marketing plugin all running at once, each with its own script.
- “All-in-one” plugins promising SEO, security, caching and backups from a single install – they usually do each of those things adequately at best, and together they drag in a lot of code that runs on every request regardless of which single feature you’re actually using.
Currency and multi-currency plugins deserve a separate mention – they often fire an external API call on every product page load to refresh exchange rates. If that call hangs, and third-party services do hang from time to time, the whole store slows down, not just the price block.
Page builders are their own story
Elementor and similar builders deserve their own paragraph, because people rarely remove them – they’ve gotten used to visual editing and worry they won’t be able to manage the site without it. Fair enough. But it’s worth knowing: every page built with one of these tools pulls its own set of CSS and JS independent of every other page, and that set is often heavier than the page actually needs.
You don’t have to ditch the builder entirely. Moving just the product page (the most visited page, where every second of delay hits conversion directly) onto a lightweight custom template while leaving the blog, the about page, and contacts on the builder as they are is usually enough.
Once you’ve found the culprit
Caching is the baseline, not a cure-all
Page caching (WP Rocket, LiteSpeed Cache, W3 Total Cache) handles some of the server-side speed issues, but it won’t save you from a heavy frontend. Cached HTML still drags along the same heavy scripts and fonts – it just gets generated faster on the server. Don’t expect a caching plugin to fix a badly written recommendations plugin on its own.
Images
On catalogs with thousands of products, images often outweigh everything else on the page combined. WebP instead of JPEG, lazy loading for anything below the fold, and – this gets missed a lot – checking the actual file size coming from the supplier’s feed. A 4000×4000 pixel image resized down to display size purely through CSS still gets downloaded by the browser at full original size.
Database cleanup
WooCommerce accumulates a fair amount of junk in the database: abandoned order drafts that never turned into orders, page revisions, payment gateway transaction logs nobody ever clears. On one store with three years of history, the postmeta table alone was over two gigabytes, and a good chunk of database queries ran through it even on pages that had nothing to do with orders. Regular cleanup (WP-Optimize, or hand-written SQL for anyone comfortable with it) genuinely takes load off.
A lighter alternative instead of the heavy plugin
Not every plugin should just be deleted – often you still need the exact feature it provides, the specific implementation just turned out to be resource-hungry. In that case it’s worth looking for a lighter alternative: a smaller plugin doing the same job, without the extra functionality bundled around it.

When a plugin is worth replacing with custom code
Sometimes no off-the-shelf plugin fits: one is too heavy, another too limited in its settings, a third conflicts with the theme. At that point it makes more sense to write a small piece of code for the specific task – a “customers also bought” block built on a custom query with proper database indexes weighs a fraction of what a general-purpose plugin does, since that plugin has to account for every possible use case, not just yours.
This isn’t always worth the budget – custom code costs more than installing a plugin. But for a block shown on every product page, to thousands of visitors a day, the speed difference usually pays for itself pretty quickly through conversions the store stops losing to slow loading.
Instead of a conclusion
There’s no fixed “always disable these five plugins” list – what slows down one store runs fine on another because it has fewer products, a different theme, or a different server setup. But the method stays the same every time: measure first, disable one plugin at a time and measure again, and only then decide whether to keep it, swap it for something lighter, or replace it with custom code. Half an hour with Query Monitor beats weeks of guessing.
Read also: How Much Does a WooCommerce Online Store Cost in Ukraine in 2026; What Is HTTP 304 Not Modified and How It Affects Site Speed.