Every Shopify store we audit has apps nobody can account for. A countdown timer from a sale two years ago. Two review apps doing the same job because someone switched and never uninstalled the first. A wishlist tool installed during a redesign that never shipped.
Each one is quietly charging a monthly fee and adding scripts to every page load. Nobody notices, because apps fail silently — they don’t break anything, they just make everything slightly slower and slightly more expensive.
This article covers how to work out which apps are earning their place, what the real cost of an app is beyond its subscription, and what to do about the ones that aren’t.
The Real Cost of an App
The monthly fee is the smallest part of it.
Page weight. Most apps inject JavaScript into every page, not just the pages where they’re used. A product review widget frequently loads on your About page, your policy pages and your blog — and page weight is one of the main causes covered in our complete guide to improving your PageSpeed score.
Leftover code. This is the one that surprises people. Uninstalling an app removes it from your admin, but code it wrote into your theme files often stays behind. We regularly find stores running scripts from apps deleted two years earlier.
Conflict risk. Two apps modifying the same part of your theme will eventually collide, usually during a sale when you’re least able to investigate.
Attention. Every app is something to configure, update and remember. Ten apps is a small system to administer.
Working Out What You Actually Use
Open Shopify Admin, then Apps, and list everything. For each one, answer three questions honestly.
When did I last open this? If the answer is “I don’t remember”, that’s information.
What would break if it disappeared tomorrow? If nothing obvious, it’s a candidate.
Could the theme do this natively? Shopify has absorbed a lot of app functionality into the platform over the past few years — and many merchants are still paying for something their theme now does.
| App category | Usually worth keeping? | Notes |
|---|---|---|
| Reviews | Yes — one of them | Check you’re not running two |
| Email and SMS flows | Yes | Highest revenue per dollar of any category |
| Buy now pay later | Yes if AOV is above ~$80 | Often native to checkout now |
| Upsell and cross-sell | Sometimes | Measure it — many add weight without lifting AOV |
| Countdown timers and urgency | Rarely | Erodes trust when the timer resets on refresh |
| Popups | Sometimes | One, configured well, not three competing |
| Page builders | Depends | Heavy, and hard to remove once pages depend on it |
| Currency converters | Usually not | Shopify Markets handles this natively |
Two categories on that list are worth keeping almost regardless of cost. Email and SMS flows produce directly attributable revenue, and reviews affect conversion at the point of decision while also feeding the star ratings in your search listings. After those two it drops off quickly.
Measure Before You Judge
The honest answer for most apps is that nobody knows whether they work, because nobody set up a way to tell.
Upsell apps are the clearest example. They’re sold on the promise of lifting average order value, which is measurable — yet most merchants have never compared AOV before and after installing one. If you can’t demonstrate the lift, you’re paying a subscription and a speed penalty for a feeling.
Before removing anything, record your baseline: average order value, conversion rate and mobile PageSpeed. Then you can tell the difference between an app that was working and one that was just present.
One caution while you’re comparing: if your conversion tracking isn’t accurate, neither is the before-and-after you’re about to run.
Removing Apps Properly
Uninstalling is only the first step, and skipping the rest is why so many stores stay slow after a clean-up.
Uninstall from the admin. Straightforward.
Check the theme files. Look for leftover script tags and snippets referencing the app. This usually needs someone comfortable in Liquid, and it’s the step that actually recovers the speed.
Check for orphaned metafields. Some apps write data to products that stays behind.
Re-measure. Compare PageSpeed before and after so you know what the removal bought you.
If you’re not comfortable editing theme code, do the uninstalls yourself and have a developer do one clean-up pass across all of them at once. It’s a far better use of a few hours than doing it app by app — and it’s the reason a slow store rarely needs rebuilding, just clearing out.
Where to Start
This week: List every installed app with its monthly cost. Add up the total — most owners are surprised. Then record your current mobile PageSpeed score.
This month: Uninstall anything you can’t justify, and have someone check the theme files for leftover code from apps you’ve removed at any point in the past.
Next 90 days: For the apps you kept, set up a way to measure whether each one is working. Anything you can’t measure by the next review is a candidate for removal. If the store is slow enough to be losing sales, the checkout diagnosis is worth running at the same time.
Frequently Asked Questions
There’s no fixed number — what matters is whether each one earns its place. That said, stores running more than about fifteen apps are usually carrying several they don’t need, and the speed cost compounds.
Only if the leftover code goes too. Uninstalling alone often produces a smaller improvement than expected, which is why merchants conclude apps weren’t the problem when they were.
No. A free app carries the same page weight and conflict risk as a paid one. It just doesn’t appear on your invoice, which makes it easier to forget.
It depends on how much custom layout work you need and whether you have development support. Be aware they’re heavy, and once landing pages depend on one, leaving is a project rather than an uninstall.
Quarterly is enough for most stores, plus a check any time you install something new. Apps accumulate the way browser tabs do.