Short answer: Go to Settings, then Billing. Copy each paid app into a spreadsheet with what it cost over the last 3 bills. Mark any app whose job the theme or another app already does. Remove those one at a time, and check the product page and cart on your phone after each removal. Reviews and analytics stay.
Below is how we'd go through it, step by step, with a spreadsheet open the whole time. We'd take out about one app a week. You won't see dollar benchmarks here; your own billing page is more accurate than anything we could print.
Step 1: where do I see what each app costs me?
In the admin. Settings, then Billing. App charges sit on your Shopify bills, and Shopify's page on app billing in Shopify shows where each one appears.
Pull the last 3 bills, not just the latest. Usage-priced apps, the ones charging per order or per email, rise and fall with your sales, and a single bill hides that.
Copy every app line into a sheet. One row each.
Step 2: what goes in the spreadsheet?
We use six columns. You can add more, but we've never needed to.
- App name and the job it was installed for.
- Monthly cost, averaged over those 3 bills.
- Where a shopper sees it (write "nowhere" if they don't).
- Does the theme or another app already do this?
- Any report that ties it to orders.
- Who asked for it, and whether they still log in.
Fill column 5 honestly. An app with attributed orders in its own report goes to the "keep" pile right away. If no one on the team can say what an app does, mark it red.
Check the install dates while you're there. Trials that rolled into paid plans tend to show up here. So do campaign apps nobody has opened since the campaign ended.
Step 3: how do I spot apps doing the same job twice?
Sort column 4. Then open the theme editor, click into a product page section, and read the full block list. Themes ship blocks for things people still pay apps for, and countdown timers and trust badges are the usual ones. If subscriptions are a big line item, check the subscription app fee math.
The overlaps we run into most:
- A pop-up app sitting next to an email platform that already has sign-up forms.
- A bundle app on the product page plus a separate upsell app in the cart.
- A page builder bought for one landing page that no longer gets visits.
- Image compression in one app and meta tags in another, plus a schema app on top, all touching the same pages as the theme.
Two bundle apps mean two sets of scripts loading for one offer. That's a speed cost on top of the bill.
For upsell tools specifically, our comparison of Kaching Bundles vs Candy Rack vs ReConvert lists where those three apps overlap.
What does a filled-in sheet look like?
Here's an illustrative one. The rows are invented to show the method, so there are no prices in them; put your own from the billing page.
- Review app, installed at launch. Shoppers see it on every product page. Its report ties it to orders. Keep.
- Countdown timer, added for a sale. The sale ended long ago, and the product page shows nothing. Red row. The theme may have a timer block anyway.
- Pop-up app. The email platform the store already pays for has its own sign-up forms. Yellow: try the built-in form on a duplicate theme first.
- Page builder. Bought for one landing page. Nobody can say when that page last got traffic. Red.
- Cart upsell app. Attributed orders show in its report and roughly match Shopify's. Keep it. Peek at the lower plan, though.
- Second bundle app. Someone installed it to test one offer, and now both bundle apps load scripts on the same product page. Red, after you confirm which one the live offer actually runs on.
Three reds, a yellow, two keeps. We'd pull the timer first since nothing on the storefront uses it. The page builder goes next. Hold off on the bundle app until you know which offer is live.
Not glamorous. It works, though, because every decision is tied to something you can check in the admin and on your phone, not a gut feeling about which app "seems" useful.
Step 4: should I swap several apps for one?
Sometimes it does. You'd pay one bill where you used to pay several, and the product page usually loads fewer scripts. But a widget bundled into a bigger app tends to do less than the specialist app it replaced. If that one vendor has a bad week, so does every widget on your store.
On speed, see our write-up on whether one all-in-one Shopify app can replace 6 widgets. It has the results.
Disclosure again, since it's relevant here. We make Nitro: Upsell & Cross-Sell on the Shopify App Store. It's free to install. On November 3, 2026, the listing showed a 5.0 rating from 254 reviews. Line it up against what you already pay for.
Before any swap, list what the old app did for your store this week. Nothing more. If the new app handles everything on that list, try it on a duplicate theme first. Whatever else it advertises shouldn't sway you.
Step 5: how do I remove an app without breaking anything?
Slowly. Here's the sequence:
- Duplicate the live theme first. Shopify's help page on duplicating themes covers it.
- Choose a single app from the red rows.
- Export its data, especially reviews and subscriber lists.
- Uninstall it from the admin using Shopify's guide to uninstalling apps.
- On your phone, open a product page, a collection, and the cart. Look for empty gaps.
- Wait a few days. Then, the next row.
Some older apps leave snippets in the theme after uninstall. Our guide on why Shopify themes slow down from leftover app code shows how to find them.
Step 6: which apps should I not touch?
Three kinds, in our audits.
- Reviews. Years of social proof, and a careless switch can lose the ones you imported.
- Any cart or upsell tool whose report shows attributed orders.
- Analytics and tracking, since without them you can't see whether the other cuts hurt.
Get your conversion rate and average order value on paper before you remove anything, a few weeks' worth. Look again after the last removal, once traffic settles. A drop points at whatever you uninstalled most recently. On reviews, weigh whether a $15 review plan is enough before you upgrade.
Conversion has plenty of other levers. We cover them in our guide to ecommerce CRO best practices for Shopify.
Four questions before each cancel
We ask these per row. If it disappeared tonight, would a shopper notice on the product page or in the cart? Then check that its revenue report roughly matches what Shopify shows, and look for a cheaper tier of the same app that still has the feature you rely on. Last, find out whether uninstalling wipes reviews, subscribers, or loyalty points that you can't get back.
We open the app's plan page next to the sheet and read each tier against column 1, the job the app was installed for. Sometimes a lower tier covers it. Then we'd downgrade instead of canceling, because a plan upgraded for a busy season tends to stay there after the season ends.
What about post-purchase offers and new stores?
Post-purchase upsells hide from audits. They run on the thank-you page, after payment, so you won't see them while clicking around product pages. Open the app's revenue report before you cut one. Our comparison of post-purchase upsells on the thank-you page covers the main ones.
New stores have the opposite problem: no data. Start with a free theme, one review app, and one upsell tool. Add apps only when a number in your reports tells you to.
How often should I redo the audit?
Every quarter. Plus once after BFCM or any big sale. To compare replacements side by side, visit our app comparisons hub.
After a big sale like BFCM, we sort the sheet by install date and start with whatever went in during sale week. Timers go first. A countdown timer added for the weekend can keep running into January, and a second pop-up added to push a code often does too.
We keep last quarter's tab and hold its column 2 against the new one. That catches a usage charge that crept up with orders, or an app someone reinstalled.
Whether the sheet has ten rows or forty, the method doesn't change. One app at a time, with a phone check after each.
Put the next review in your calendar today. A quarter goes by faster than you'd think.
What if my team disagrees about an app?
It happens. Someone on the team installed the app, likes it, and doesn't want it gone.
Look at column 5 together. Attributed orders in the report? Keep it. Would a shopper miss it? Also keep. Otherwise, suggest pulling it from a duplicate theme for a few days and comparing conversion rate and average order value. Let those numbers settle the argument.
That usually ends it.
Frequently asked questions
How much does a typical Shopify app stack cost per month?
We don't publish an average, because it varies too much. A five-product store and a multi-brand shop with thousands of SKUs pay very different amounts. Some apps are free, some flat-rate, and some bill per order or per email. Average your last 3 Shopify bills instead. That's the figure to cut from.
Which Shopify apps are usually redundant?
Pop-up apps, when the email platform already has sign-up forms. A bundle app running alongside a separate cart upsell app. Page builders still billing for one old landing page. And timer or trust badge apps on a theme that ships those blocks for free.
How do I uninstall a Shopify app without breaking my store?
Duplicate your live theme first, so you have a rollback. Export any data you need, such as reviews. Remove one app at a time from the admin, then check product and cart pages on a phone. Look for leftover code in the theme, and wait a few days before the next removal.
Can one all-in-one app replace several Shopify apps?
It can. You pay one subscription instead of several and usually load fewer scripts. Widgets inside an all-in-one app tend to be simpler than the specialist apps, and you depend more on one vendor. Write down what you used the old apps for this week, and check the replacement against that list.
Which apps should I keep to protect conversion rate?
We'd leave the review app alone first, because reviews take years to collect and a careless switch can lose imported ones. Cart or upsell apps whose own reports show attributed orders come next. Keep analytics. Without it, there's no way to see whether a cut hurt sales.