Which Shopify apps are essential?

No app is strictly essential to run a Shopify store: the platform already handles products, orders and most of payment natively. Apps become useful for precise needs: reviews certified by a third party, advanced automated email marketing, or features specific to a sector. But building blocks like filters, upsells, the announcement bar or a simple review module can be coded straight into the theme instead of handed off to a subscription. The rule we apply: an app is justified when it brings a service that cannot be rebuilt cleanly in-house, like a certified payment gateway, never for a feature that fits inside a theme section. Before installing a new app, ask whether the feature could be coded once and for all instead of rented every month.

How much does a Shopify store cost when it's built without piling up paid apps?

A Shopify store built with features coded straight into the theme instead of added through apps costs from €3,000 excl. VAT, for a new build or a redesign. Allow 4 to 6 weeks, for a new build as for a redesign. The exact quote is set as a fixed fee after a discovery call, based on the number of features to code and the size of the catalog: a redesign with a handful of custom features and a small catalog sits close to the floor, a full build with dozens of variants and several markets costs more. This price covers the code: once the store is delivered, features coded into the theme cost nothing more to run. Monthly support remains available after launch, with no commitment, priced according to what the store needs once it is live.

What happens if I can no longer make changes to my store after delivery?

One team and one point of contact follow the project from the first meeting to launch, and every client gets training toward independence before delivery: changing text, photos, colors and theme blocks does not require going back to the team for every content change. After launch, a bug gets a response within one business day, and deeper changes (a new section, a new feature) are handled by fixed quote or by ticket, whichever suits the brand. It is one more reason to code straight into the theme: the configuration stays in the standard Shopify admin (sections, blocks, metafields) that the client's team already knows. A separate app interface depends on its publisher, who can stop the service or raise the price.

Four to six weeks for a redesign: is that enough for custom code?

Four to six weeks is enough for a Shopify redesign because of where the time goes. Part of that time, often lost elsewhere configuring seven or eight different apps, their settings and their compatibility with the theme, gets reinvested directly into coding the features: reviews, filters, upsells, announcement bar. It shows at launch: a lower mobile LCP, fewer third-party scripts loaded, a shorter perceived load time. The precise schedule depends on the number of features to code and the size of the catalog to migrate; it gets set after the discovery call and held throughout the project, with regular check-ins along the way.

Should you still keep an app like Judge.me or Klaviyo?

Yes, though the logic is different for each. Judge.me has value when the goal is certified third-party proof: the review is verified and hosted by a platform independent from the brand, which carries more weight with a visitor than a review simply displayed on the site. Klaviyo, or any advanced email marketing tool, handles automated scenarios, segmentation and deliverability that it would not be reasonable to rebuild in-house for a store. These are two examples where the app brings a specialized service that cannot be coded cleanly into a theme. The rule still holds though: a collection filter, an announcement bar or a simple upsell module need neither of these two tools to work well, and cost zero euros a month once coded.

Shopify without paid apps: what we code straight into the theme

Seven active apps on a Shopify store means seven monthly subscriptions and seven scripts slowing it down. This guide lists what we code straight into the theme instead, and what it changes for speed and budget.

The short answer

A Shopify store without paid apps means common features (product reviews, filters, upsells, announcement bar, editable sections) are coded straight into the theme instead of handed off to a monthly subscription. That removes one network call and one JavaScript block per app, which lightens the page and removes the subscription that came with it. Some apps stay justified (payment, email marketing like Klaviyo, reviews with third-party proof like Judge.me), never for a simple theme section. At Grillé, the sections are hand-coded into the theme and the team updates its pages and catalog on its own.

Why agencies pile up apps instead of coding

A founder opens the Shopify admin one morning and counts: one app for reviews, one for upsells, one for pop-ups, a third-party collection filter, a trust badge. That adds up to seven apps and seven subscriptions, on a store that takes several seconds to load on mobile. That is what we find with most brands that reach out to us after a first experience elsewhere.

Installing a review app takes five minutes; coding an equivalent system into the theme, with a form, moderation and a clean display, takes several days. When an agency bills a tight fixed fee or works no-code, the paid app becomes the logical shortcut, and the cost moves to the client, who pays it month after month as a subscription.

The choice is made upfront: either we code it once, custom, and the feature belongs to the store for good, or the brand rents a third-party block and stays dependent on its price and its compatibility with every Shopify update. If your store drags, that slowdown probably has a precise cause.

What we code straight into the theme, feature by feature

Most Shopify themes do not natively include product reviews, advanced filters or upsells, hence the reflex to install an app. We start by looking at what Shopify already allows, then code into the theme what is missing, with data stored in metafields and no external script to load.

In every case, the logic stays the same: the feature already exists in Shopify's native capabilities, or can be coded into the theme with the same result, without depending on a third-party service billed every month. The cases below come up on most of the Shopify projects we deliver.

Product reviews: a module coded into the theme, data stored in Shopify metafields, indexable by Google without an external script blocking the display.
Collection filters: Shopify's native capabilities (Search & Discovery) completed with an interface coded into the theme, without a separate rendering script.
Upsells and bundles: complementary products and grouped offers coded into the cart and the product page, with the rules managed in the theme.
Announcement bar and pop-ups: a banner or email capture form coded into the theme, without the several-hundred-kilobyte third-party script that pop-up apps load.
Editable sections: every coded block stays editable from the Shopify admin, text, photos, rules, without a separate app interface for the smallest change.

What is the impact on speed and monthly cost?

Every installed app adds a network call, sometimes several, and often a JavaScript block that runs before the page becomes interactive. Stacked on top of each other, these apps delay the display of the main content (the LCP), a delay that weighs on conversion and on Google's ranking, since Google takes page experience into account in its ranking.

On the budget side, each app is billed monthly, at the price its developer sets and lists on the Shopify App Store. Once coded into the theme, the same feature costs nothing to run.

At Grillé, the team adds a new book, builds a campaign page or reorders the home page on its own, with no app and no developer.

	Paid app	Coded into the theme
Monthly cost	One subscription per feature, at the developer's price	No subscription once the store is delivered
Scripts loaded	One network call and one script per installed app	No third-party script
Who is in control	The app's publisher, at their price and pace	The brand, through the theme's sections

Which apps are still worth paying for, and why

We do not code everything. Some services require an infrastructure that would be neither reasonable nor safe to rebuild in-house. Payment is one of them: payment gateways are certified, audited, and continuously updated on security matters that go beyond the scope of a theme.

Advanced email marketing tools like Klaviyo are another: building a comparable scenario and deliverability engine would take months for a less reliable result. Same logic for reviews when the goal is certified third-party proof like Judge.me: the value comes from the review being verified and hosted by an independent third party. The rule: an app is justified when it brings a service that cannot be coded cleanly, never for a simple theme section.

What this changes for independence after launch

One side effect of piling up apps: each one has its own admin interface, its own password, separate from the standard Shopify back office. The founder has to learn seven different tools to run the store, and the smallest change to upsell logic means reopening the documentation for that app.

With code built into the theme, the configuration stays in the Shopify admin the team already knows: sections, blocks, metafields. We train each client to be independent on their own theme before delivery, to avoid this dependency on third-party interfaces. A bug after launch gets a response within one business day, and further changes are handled by fixed quote or by ticket.

What this means for your next Shopify project

Two paths exist for launching or rebuilding a Shopify store. The first: start from a theme and pile up apps as needs come up, accepting a monthly subscription that grows with every new feature. The second: decide upfront what should be coded into the theme (reviews, filters, upsells, announcement bar) so the store stays light and the cost stops once the project is delivered.

A small store testing a market may prefer the speed of an app, and migrate later if needed. A brand that already has traffic should look at what, among its current apps, could be coded once and for all. A Shopify redesign built this way takes 4 to 6 weeks, with a quote set after a discovery call. Details on /services/e-commerce.

Frequently asked questions

Which Shopify apps are essential?: No app is strictly essential to run a Shopify store: the platform already handles products, orders and most of payment natively. Apps become useful for precise needs: reviews certified by a third party, advanced automated email marketing, or features specific to a sector. But building blocks like filters, upsells, the announcement bar or a simple review module can be coded straight into the theme instead of handed off to a subscription. The rule we apply: an app is justified when it brings a service that cannot be rebuilt cleanly in-house, like a certified payment gateway, never for a feature that fits inside a theme section. Before installing a new app, ask whether the feature could be coded once and for all instead of rented every month.
How much does a Shopify store cost when it's built without piling up paid apps?: A Shopify store built with features coded straight into the theme instead of added through apps costs from €3,000 excl. VAT, for a new build or a redesign. Allow 4 to 6 weeks, for a new build as for a redesign. The exact quote is set as a fixed fee after a discovery call, based on the number of features to code and the size of the catalog: a redesign with a handful of custom features and a small catalog sits close to the floor, a full build with dozens of variants and several markets costs more. This price covers the code: once the store is delivered, features coded into the theme cost nothing more to run. Monthly support remains available after launch, with no commitment, priced according to what the store needs once it is live.
What happens if I can no longer make changes to my store after delivery?: One team and one point of contact follow the project from the first meeting to launch, and every client gets training toward independence before delivery: changing text, photos, colors and theme blocks does not require going back to the team for every content change. After launch, a bug gets a response within one business day, and deeper changes (a new section, a new feature) are handled by fixed quote or by ticket, whichever suits the brand. It is one more reason to code straight into the theme: the configuration stays in the standard Shopify admin (sections, blocks, metafields) that the client's team already knows. A separate app interface depends on its publisher, who can stop the service or raise the price.
Four to six weeks for a redesign: is that enough for custom code?: Four to six weeks is enough for a Shopify redesign because of where the time goes. Part of that time, often lost elsewhere configuring seven or eight different apps, their settings and their compatibility with the theme, gets reinvested directly into coding the features: reviews, filters, upsells, announcement bar. It shows at launch: a lower mobile LCP, fewer third-party scripts loaded, a shorter perceived load time. The precise schedule depends on the number of features to code and the size of the catalog to migrate; it gets set after the discovery call and held throughout the project, with regular check-ins along the way.
Should you still keep an app like Judge.me or Klaviyo?: Yes, though the logic is different for each. Judge.me has value when the goal is certified third-party proof: the review is verified and hosted by a platform independent from the brand, which carries more weight with a visitor than a review simply displayed on the site. Klaviyo, or any advanced email marketing tool, handles automated scenarios, segmentation and deliverability that it would not be reasonable to rebuild in-house for a store. These are two examples where the app brings a specialized service that cannot be coded cleanly into a theme. The rule still holds though: a collection filter, an announcement bar or a simple upsell module need neither of these two tools to work well, and cost zero euros a month once coded.