Analyze with AI

Get AI-powered insights from this Mad Devs article:

Imagine a Shopify store where customers can buy individual products, but cannot yet build their own discounted bundle. The team wants shoppers to choose three different eligible items and receive a discount. Its merchandiser must also change the eligible assortment without waiting for a frontend release.

At the planning meeting, someone proposes a separate storefront. Before anyone estimates that rebuild, the team has a practical question: which layer handles each requirement – product selection, bundle eligibility, discount calculation, checkout, and assortment updates?

This guide compares headless and traditional commerce through three Shopify projects, then shows how to turn that buying flow into a testable scope of work.

Key takeaways

  • Choose the architecture after checking where a specific store feature is blocked. A difficult interface, an incompatible app integration, and an unsupported selling rule require different responses.
  • Moving the interface and changing the rules of sale create separate work. A new storefront still needs a confirmed way to handle the offer through checkout.
  • Request comparable estimates using one buying scenario and identical acceptance criteria. Include the merchandiser's work after launch, alongside the customer's path to purchase and the engineering team's responsibilities.

What changes when you separate the storefront

A storefront built on the platform’s theme

With a Shopify theme, the customer interface sits within the platform's theme system. Features can combine theme code, platform capabilities, and apps. Shopify's theme app extensions let merchants place and configure app blocks through the theme editor.

The block is an interface entry point. Its presence does not establish where the app enforces a discount or manages eligible products. During evaluation, ask the implementer to identify the theme component and the service responsible for the selling rule. Those dependencies determine what must change. Record those dependencies before assuming that replacing the visible block will preserve the feature.

A separate storefront connected through APIs

A headless storefront is a separate frontend application that obtains commerce data and capabilities through APIs. The existing commerce backend can remain in place. Content may also come from a headless CMS, which supplies managed content through an API.

Separation does not require replacing checkout. In Shopify's documented flow, the frontend queries the cart's checkoutUrl when the buyer is ready and sends them to Shopify Checkout. This Storefront API cart guide describes the hosted checkout handoff, checked November 20, 2026.

The team must still verify how each required app connects to the separate frontend.

Why three Shopify merchants made different storefront decisions

Luxury For You moved to a separate storefront

Luxury For You wanted a distinctive catalog experience, brand navigation, and custom merchandising. According to Stackframes' project account, the contractor replaced the standard storefront with a Next.js frontend connected to Shopify's Storefront API, while retaining Shopify as the commerce backend.

The delivered scope included filterable category pages, navigation across designer brands, member pricing, and editorial shopping experiences. These functions describe what Stackframes built. The case does not establish that every individual function required headless architecture or was impossible in a theme.

For another merchant, the useful question is which mandatory interactions exceed its current implementation's capabilities. A request for more control becomes estimable when the team can identify the screens, data, and behavior requiring that control.

Katalyst moved back to a Shopify theme

Katalyst wanted easier maintenance and updates alongside storefront improvements. Lazer's case study describes replacing its Builder.io headless frontend with a Shopify Liquid theme. The contractor organized content into reusable sections and blocks, then trained the merchant's team to manage it.

The purchase flow also connected a physical workout suit with a Recharge subscription. That required integration work beyond recreating page layouts. The theme supplied the storefront experience; the subscription connection remained a distinct implementation responsibility. This project does not demonstrate a capability of the standard Shopify Bundles app.

The operational requirement matters: who can publish the next content change, using which controls? For stores moving managed content between systems, planning a CMS migration is a related scope task that should be identified explicitly.

Parachute revisited its original decision

Shopify's Parachute account describes a headless decision in 2019, driven by complex bundle selling and growing omnichannel requirements. By 2024, the brand chose to return to Shopify's native capabilities as the platform evolved and maintaining custom functionality became an operational concern.

The account also identifies native split shipments as a reason to reconsider the custom setup. The lesson for a new estimate is to retest the original constraint against the current platform and operating requirements. A past limitation is evidence about that earlier decision, not proof that a rebuild remains necessary today.

MerchantReason for changeWhat the team changedQuestion to check in your store
Luxury For YouCustom catalog, navigation, and merchandising requirementsStackframes built a Next.js storefront over ShopifyWhich required interactions need a separate frontend?
KatalystSimpler content updates and maintenanceLazer rebuilt the storefront as a Liquid theme and integrated RechargeCan the team publish changes and preserve the purchase integration?
ParachuteEvolving platform capabilities and operational needsThe brand returned to native Shopify capabilitiesDoes the original constraint still apply?

These projects support different scopes. Use their questions to investigate your store before treating an architectural direction as a project requirement.

Find what limits one buying flow in your store

Define the offer and the expected outcome

Return to the hypothetical Shopify store. One complete bundle contains three different eligible products, one unit of each. Products A, B, and C cost $24, $20, and $16. The offer gives 10% off only when the bundle is complete. All amounts below are USD, before shipping and taxes; these are assumptions for this example.

Write the expected results before evaluating either frontend option. The same conditions must apply to both implementations, including changes made after the shopper starts selecting products. Keep the offer definition with the demonstration so everyone evaluates the same behavior.

ScenarioExpected result
The shopper selects A, B, and CSubtotal $60, discount $6, subtotal after discount $54.
The shopper removes CA and B remain at $44. The bundle discount disappears.
C becomes unavailable before checkoutThe team verifies that checkout rejects a bundle containing the unavailable item and the shopper can choose a replacement.
The merchandiser removes C from the offerThe updated assortment applies to new selections without a frontend release.

The availability check reaches beyond the selection screen. A convincing builder demo is insufficient if the current checkout still accepts an invalid combination. Likewise, a technically correct purchase flow fails the business requirement if every assortment update needs an engineer.

Identify who implements each requirement

Assign the work across three parts of the store:

  • Theme or storefront application: the selection interface and the message explaining an unavailable item. Engineers demonstrate how the shopper recovers and chooses a replacement.
  • Commerce platform or bundle app: eligible composition, discount enforcement, and cart handling through checkout. The offer owner confirms the rule, and the implementer demonstrates it in the store's configuration.
  • Operational process: the place where the merchandiser changes eligibility and how published changes reach shoppers. The merchandiser performs the update during acceptance.

As checked November 20, 2026, Shopify's bundle documentation distinguishes fixed bundles and multipacks, supported by Shopify Bundles, from mix-and-match offers. For mix-and-match, it describes third-party apps or a custom offering through Bundles APIs on Shopify Plus. Shopify Bundles itself does not support mix-and-match.

First confirm that the chosen configuration supports the selling rule. Then evaluate the interface. Changing the frontend alone cannot establish a missing commerce capability, and a theme app block does not prove that the same app exposes a usable headless integration.

Turn the findings into two comparable scopes

Specify the work for each option

Give both estimators the offer definition, four acceptance scenarios, current checkout, and merchandiser workflow. Ask them to identify confirmed capabilities, unresolved dependencies, and who will maintain each changed component. This makes differences in the proposals visible before comparing totals.

Theme development. Suppose the theme supports the required placement and a verified app implements the mix-and-match rule. The scope covers styling and assembling the builder, configuring the offer, checking the four scenarios, and documenting assortment updates. Confirm which controls remain available in the theme editor and which belong to the app. Include the merchandiser's acceptance in the delivery criteria.

Separate storefront. The estimate needs a custom bundle interface, verification of the app's integration with the independent frontend, cart integration, and the handoff to the existing checkout. Ask for evidence that the app can supply the required behavior outside the theme. Include deployment and ongoing maintenance of the new application. The four acceptance scenarios remain identical; a different stack does not change the promised offer.

Shared dependency. If the configuration cannot enforce the discount or bundle composition, estimate the commerce-logic change separately. Resolve that question before estimating a frontend transition. Otherwise, one proposal may assume an available app while another includes custom selling logic, leaving the totals incomparable.

For either option, request a clear list of deliverables and exclusions. An estimate should say whether integration validation is complete or still an investigation, and whether merchant training and release documentation are included. Unconfirmed app behavior needs an explicit verification task and an owner.

Resolve the example and define the handover

In this hypothetical scenario, the team tests a suitable app on the existing theme. It supports the bundle rules, applies the discount correctly in the current checkout, and lets the merchandiser update the assortment without a frontend release. The required customer interface fits within the theme's capabilities. These are stipulated findings for the example, not results attributed to a named app, Shopify, or Mad Devs.

The store chooses theme development. Its scope is to configure the offer, build the selection interface in the theme, verify the four agreed scenarios, and document how to change the assortment.

The engineering team owns the frontend implementation. The offer owner approves the discount rule and eligible product list. The merchandiser confirms that they can publish an assortment change independently. Handover includes the configuration, changed code, acceptance evidence, and operating instructions, so the next update has a clear owner.

Revisit the architecture when a mandatory customer scenario cannot be delivered through the current theme and compatible apps, while the commerce platform's purchase rules are confirmed. That creates a concrete reason to estimate a separate storefront. Until then, the verified scope addresses the requirement that started the discussion.

Conclusion

Write down the requirement that prompted the storefront rebuild discussion, then request confirmation of where each mandatory outcome is delivered. Include customer interaction, selling rules, checkout behavior, and the merchandiser's update process in that record. Ask both implementers to estimate against it and identify any remaining assumptions. That record gives the decision a practical basis: a defined buying flow, evidence about the current configuration, and a scope whose completion the team can verify.

Frontend development
Upgrade your storefront on your existing platform
Mad Devs can help you define what your storefront needs, then improve the existing frontend or build a separate application on your commerce APIs. We can validate the buying flow and hand over the code, integrations, and release documentation to your team.
Discuss your storefront

Latest articles here

Two payment gateways connected through a routed transaction flow

Multiple Payment Gateway Integration: How to Add a Second Provider

Your store is preparing to sell in another market, but its existing payment connection does not meet the new checkout requirements. You already need...

AI analysis core comparing workflow candidates to prioritize an AI use case.

AI Opportunity Assessment: How to Choose Which AI Use Case to Build First

Leadership wants AI on the product roadmap. Customers have suggested a copilot, support wants less manual work, and engineering has ideas for...

Conveyor motor and bearing visualization for predictive maintenance with machine learning

Predictive Maintenance With Machine Learning: Choosing a Failure Target and Validating Early Warnings

An industrial software provider wants to add failure warnings to equipment dashboards that already collect telemetry. Its CTO and the customer's...

Go to blog