The Shopify Plus app-stack audit: cut the double-pays

Most Shopify Plus brands have never run a line-by-line audit of what they pay for apps. The stack grows the way a junk drawer does. A problem shows up, you install a tool to solve it, and six months later you install a different tool that does the same thing, but you never cancel the first one. At 15 to 30 active paid apps, which is normal for a mature Plus store, the annual bill quietly reaches $12,000 to $43,000 or more. This is the 90-minute audit we run with operators to find the double-pays, the six categories where they hide, and how to cut without breaking checkout.

The $43,000 problem: how an 18-app stack happens

The all-merchant average is about 6 apps per store. Shopify Plus brands routinely run 15 to 30. That gap is not a strategy. It is the compound result of solving problems one at a time and never subtracting.

Here is the pattern we see again and again. A brand adds a reviews app in year one. In year two the marketing hire prefers a different reviews app and installs it, but nobody uninstalls the first. Add a popup tool, then an email platform that also does popups, then an analytics app the agency recommended, then a second analytics app after the first one's attribution looked wrong. Each decision was reasonable on its own. Nobody ever sat down with the full list.

The dollars follow the app count. A brand under $1M in GMV typically spends around $2,160 a year on app subscriptions. By the $5M to $15M tier that figure runs to roughly $24,000 at the middle and $43,000 at the high end, and it keeps climbing from there.

Notice the size inversion underneath these numbers. As a share of revenue, smaller brands overpay: DTC tech spend runs 4 to 7% of revenue at sub-$3M brands and compresses to 0.25 to 1% at $50M+. A healthy mid-market brand targets 2 to 4% of revenue on platform, apps, and payments combined. If you are a $5M brand paying $43,000 for apps alone, you are near the top of that band before you have added the Shopify Plus platform fee, payment processing, or your agency retainer.

When I talk to founders running a brand this size, the reaction to the full billing export is almost always the same. It is not "which of these should I cut," it is "wait, we still pay for that?" The audit's real job is to force the founder to look at a list they already suspect is bloated.

The 6 overlap zones: where brands double-pay every month

Redundancy in a Shopify stack is not random. It clusters in six categories where two tools plausibly both earn a place until you look closely, and then one of them clearly does not. Across audits, 20 to 30% of an installed stack turns out to be redundant or unused, and these six are where most of that lives.

Category	Typical overlap	Annual savings (cut 1 of 2)	Common tools that collide
Email + SMS	A dedicated SMS tool running alongside an email platform that already sends SMS	$3,000-$12,000	Klaviyo, Postscript, Privy, Omnisend
Product reviews	Two review apps collecting and displaying the same reviews	$1,200-$6,000	Judge.me, Loox, Yotpo, Okendo, Stamped
Loyalty	A point-solution loyalty app plus a loyalty feature already bundled in another tool	$1,800-$8,400	Smile.io, LoyaltyLion, Yotpo, Growave
Analytics + attribution	Two attribution tools pointed at the same order data, disagreeing	$3,600-$18,000	Triple Whale, Northbeam, Lifetimely, Peel
Upsell + cart	Three upsell widgets stacked in the cart and post-purchase flow	$1,200-$4,800	ReConvert, Aftersell, Honeycomb, CartHook
Popup + onsite	A standalone popup app plus popups your email tool already ships	$1,200-$5,400	Privy, OptiMonk, Justuno, Klaviyo popups

The category that hurts most on the dollar line is analytics and attribution, because those tools are priced high and brands most often run two of them at once. The pattern we hear from operators is that the second attribution tool got bought to check the first one, and then both kept billing while the team argued about which number to trust.

There is a human version of this too. One operator, reviewing a Triple Whale line, asked plainly: "With Alana leaving, how much use is that getting?" That is the whole story of analytics bloat. A tool gets bought for one person on the team, that person leaves, and nobody cancels the seat because nobody remembers it is theirs to cancel.

The 90-minute app audit: a step-by-step template

You do not need a consultant to run the first pass. You need 90 minutes and a willingness to open the billing export. Here is the template we use.

Step 1: Pull the full 12-month spend (20 minutes). Export your Shopify billing history for the trailing 12 months. Then, and this is the step everyone skips, pull card and bank statements and find every app that bills outside the Shopify admin. Attentive, Gorgias, and many enterprise tools charge you directly, so they never appear on the Shopify bill. Build one list with a monthly and annualized figure for every tool.

Step 2: Map each tool to one business outcome (25 minutes). Next to each app, write the single outcome it exists to produce. "Collects and displays reviews." "Sends abandoned-cart email." If you cannot name the outcome, or two apps produce the same one, flag the row. This is where the six overlap zones surface on your own stack.

Step 3: Apply TIME triage (30 minutes). Sort every tool into one of four buckets. This is Gartner's TIME model, and it maps cleanly onto a Shopify stack.

Bucket	What it means	Shopify example	Action
Tolerate	Low value, but cheap and low-effort to keep	A legacy popup app barely used	Leave it this quarter
Invest	High value, under-used features you already pay for	An email tool with under 30% of flows live	Train the team before buying more
Migrate	High value, wrong tool for the price	A $600/mo reviews app where a $15/mo one covers the job	Plan a replacement with data export
Eliminate	Low value and redundant or unused	A second email platform, a third analytics tool	Cancel after exporting data

Step 4: Rank the cuts by risk (15 minutes). Do not sort your Eliminate list by dollar value. Sort it by how close each app sits to the money. A redundant reviews widget is a safe same-day cut. Anything touching checkout, live email flows, or active subscriptions goes into a tested-removal queue, not a cancel-now queue. We will get to why in the last section.

When we have struggled with an audit that stalled, the fix was almost always making it a single 90-minute block on the calendar with the billing export already open. Left as a background task, it never gets done, and the double-pays keep billing.

Reading the real number: visible fees vs. total cost

Even a clean 90-minute audit can understate your spend, because the Shopify billing page shows you only one of four cost layers. For one $1M-MRR brand that ran a full accounting, visible subscription fees were $4,500 a month. The true stack cost was $20,000 a month. The other 77% was usage fees, per-order charges, and integration and maintenance the billing page never surfaces.

Treat this one brand as an illustration, not an average, because the hidden-cost share swings hard by app category. But the layers are real and worth checking on your own stack:

• Usage fees scale with something you do not control at signup: contact count for email, session volume for analytics. The $200/month email plan becomes $900 once your list grows.
• Per-order fees are the sneakiest. One operator put it exactly right: "anything, if any order runs through a subscription app, they're taking something, usually, unless you've negotiated it differently." Subscription and upsell apps often skim a percentage of each order, and that line does not appear as an app subscription anywhere.
• Integration and maintenance is the developer time to keep the apps talking to each other. Every added app is a connection that can break during a theme update.

There is a performance tax on top of the dollar cost. Shopify's own developer documentation names installed apps as one of the biggest factors affecting web performance, because each one adds JavaScript and CSS to every page load. Worse, leftover code from apps you already uninstalled keeps dragging your speed until someone removes it by hand. A bloated stack is not just an expensive line on the P&L. It is a slower store, and a slower store converts worse.

How to cut without breaking things

The audit tells you what to cut. The execution decides whether the cut costs you revenue. Big-bang removals, where a team cancels a batch of apps in one afternoon, break something a meaningful share of the time. Sequence the work instead.

Export first, always. Before you cancel anything, export its data: review content, loyalty point balances, email flows and templates, customer segments. Once you cancel, that data can be gone in days, and re-collecting reviews or rebuilding a segment is far more expensive than the subscription you saved.

Test removal in a staging theme. Duplicate your live theme and uninstall the app there first. Apps inject code into your theme, and some leave orphaned snippets that only surface as a broken cart or a missing widget after the app is gone. Catch that in staging, not in production during a sale.

Redirect anything the app owned. If an app created URLs, a landing page, a loyalty portal, a reviews page, set up 301 redirects before you cancel so you do not lose the SEO equity or hand a customer a dead link.

Cut redundant before you cut anything load-bearing. Start with the second tool in an overlap category, because you already have the first covering the job. Save the migrations, moving off an overpriced tool to a cheaper equal, for after the easy wins prove the process works.

The theme underneath all of this is one an operator named for us after a complexity review: "the crux of it was simplification." The audit is a forcing function, and it is the kind of line-by-line P&L work our fractional CFO services run with brands every month. It makes you act on the bloat you already know is there. And once someone opens the billing export, as another operator said while working through their own list, "we can delete this, we can delete that." The candidates are obvious. The only hard part is sitting down to look.

Nobody builds a bloated app stack on purpose. It accrues one reasonable decision at a time, and the replaced tool never gets canceled. The audit is not a clever framework. It is 90 minutes with the billing export open, a column for what each app actually does, and the discipline to cut the double-pays by risk instead of by fear.

Related reading. For what brands at your size actually spend on apps, see Shopify app spend by revenue band.

Sources and methodology

App-cost benchmarks by revenue tier. Annual app-subscription ranges were compiled from 2026 Shopify Plus pricing guides and a total-cost-of-ownership model for a $10M GMV brand, cross-checked against store-level marketplace subscription data. See IWD Agency's Shopify Plus pricing analysis for the mid-market TCO model that anchors the $5M-$15M tier.

The hidden-cost composition. The four-layer split of visible subscription fees versus usage, per-order, and integration costs comes from a published 2026 audit of a single $1M-MRR Shopify brand. It is one brand's accounting, presented as illustrative rather than as an average. See App Store Research's Shopify app stack cost audit.

Consolidation frequency and the six overlap zones. The categories most often consolidated, the 20-30% redundancy finding, and the caution against big-bang removals are drawn from a 2026 stack-consolidation playbook. See App Store Research's consolidation playbook (published Apr 29, 2026).

SaaS underutilization. The 46% underutilized-or-unused figure is from a 2026 SaaS management benchmark covering application spend and usage across organizations. See Zylo's analysis of unused software.

The TIME triage model. The Tolerate/Invest/Migrate/Eliminate framework is Gartner's application-rationalization model, summarized in the LeanIX TIME model reference. We map it onto a Shopify app stack rather than a general IT portfolio.

App performance impact. The claim that installed apps are a leading factor in store speed is from Shopify's own web-performance documentation, which also notes that leftover code from uninstalled apps persists until removed.

Operator observations throughout are drawn from anonymized ecommerce founder conversations and are paraphrased to remove any identifying detail. Dollar figures attributed to named brand outcomes bundle app consolidation with other operational changes and should be read as directional, not as clean attribution to a stack audit alone.

Frequently asked questions

how do i know if my shopify app stack is too expensive?

Add every app charge across a trailing 12 months, including anything billed outside the Shopify admin, then divide by revenue. If platform, apps, and payments together run above roughly 4% of revenue at $5M+, or you can't name the business outcome each app drives, the stack is too expensive for its size.

what's the average number of shopify apps a plus brand runs?

The all-merchant average is about 6 apps per store. Shopify Plus brands routinely run 15-30 because they add tools as they scale and rarely remove the ones they replace. There is no public Plus-only figure from Shopify, so that range comes from practitioner audits, not a disclosure.

which shopify app categories overlap the most and can be consolidated?

Six: email and SMS, product reviews, loyalty, analytics and attribution, upsell and cart, and popups. These are the categories where brands most often run two tools that do the same job, so cutting one rarely costs any capability.

how do i find out what i'm actually paying in total for my shopify apps?

The Shopify billing page only shows subscription fees. Pull the full billing export, then check card and bank statements for apps billed outside Shopify, and add usage-based tiers (contact count, order volume) and any integration or dev retainer. That total is often well above the subscription line.

which apps can i safely cancel without hurting revenue?

The safest cuts are the redundant ones: a second tool in a category where a first tool already covers the job, and tools with no active usage. Cancel those after exporting their data. Anything touching checkout, email flows, or live subscriptions should be tested in staging before removal, not cut cold.

what should my tech spend be as a percentage of revenue if i'm a $5m dtc brand?

A healthy mid-market brand targets 2-4% of revenue on platform, apps, and payments combined. Sub-$3M brands often overpay at 4-7%, and $50M+ brands run at 0.25-1%. If you're a $5M brand above 4%, consolidation should come before your next scaling push.

do shopify apps slow down my store and hurt conversion?

Yes, they can. Shopify's own developer docs call installed apps one of the biggest factors affecting web performance, because each one adds JavaScript and CSS to your pages. Leftover code from uninstalled apps keeps affecting speed until it's manually removed, so cleanup is part of the audit, not just cancellation.

what is the gartner time model and how does it apply to shopify apps?

TIME sorts every tool into Tolerate, Invest, Migrate, or Eliminate. For a Shopify stack: Tolerate the cheap low-value app, Invest by using more of a tool you already pay for, Migrate off an overpriced tool to a cheaper equal, and Eliminate the redundant or unused ones. The Eliminate bucket produces most of the dollar savings.