Unifying Releases Across Eight Shopify Stores for Bears with Benefits

Bears with Benefits is a well-known German brand of vitamins and dietary supplements in gummy form. According to the brand, more than 500,000 customers have purchased its products, and its average rating on the website is 4.8/5. The products are sold through its own online stores, at dm, Rossmann, and Douglas, and in pharmacies.

Online sales run through eight separate Shopify stores serving Germany, France, Italy, Spain, the United Kingdom, the United States, Switzerland, and the Netherlands. Each has its own GitHub repository and theme, and Italy also has a separate development store. The architecture itself worked well. The problem was the process: before we joined, a separate theme was created to test each new feature, and some changes went into production without proper QA.

When the client lost its lead developer, we were brought in as a temporary replacement. The collaboration did not remain temporary: we now provide ongoing technical support for the brand’s Shopify stores.

There was no reason to rebuild the stores: the architecture worked; the release process did not. So we started with how changes reached production while also taking on tasks that existing tools could not handle.

• Unified release process: Introduced the same testing and release procedure across all eight stores, replacing the practice of creating a separate theme for each feature and deploying changes without QA.

• Custom discount app: Existing solutions could not support the required promotions, so we began developing a custom Shopify app using Shopify Functions, Shopify’s native mechanism for custom logic, including discounts.

• Pickup point addresses in emails: In several stores, pickup point details come from two sources. Bigblue writes them to an order metafield, while a custom courier integration uses its own fields. We built logic that handles both formats. This allows both integrations to work at the same time and ensures customers receive the pickup point address in their confirmation email.

• Structured data and CRO: Configured Schema Markup for product data. Together with an external CRO agency, we are implementing functional and visual changes to improve conversion rates across the Shopify stores.

From off-the-shelf solutions to a custom app

We started with the obvious option: Shopify’s standard discounts. When those were not enough, we tried existing apps such as Rebuy. Neither could support the required promotions, making custom discount logic the most complex part of the project. We therefore built a Shopify app using Shopify Functions. The discount logic lives in the app and runs directly in the cart and at checkout.

Integrations and data handling

• Pickup points. For several stores, we configured custom carrier pickup points and logic that handles two data formats: Bigblue’s order metafield and the fields used by a custom courier integration. The logic recognizes both formats and adds the pickup point address to the confirmation email. Both integrations can run in parallel without interfering with each other.

• Structured data. We configured Schema Markup so product data is available to search engines and other services in a structured format.

Tech stack

• Liquid

• Shopify Functions

• Custom Shopify Apps

• Schema Markup / Structured Data

• Changes are tested before release. Across all eight stores, new features now follow the same testing process instead of reaching a live site from a separate theme without QA.

• Promotions are no longer limited by standard discounts. The custom app supports promotions that neither Shopify’s standard discounts nor existing apps could handle.

• Customers know where to collect their orders. In stores that offer pickup point delivery, the address appears in the confirmation email regardless of which integration handles the delivery.

• The CRO agency’s recommendations reach the live stores. We implement visual and functional changes in close collaboration with the agency.

The work continues: alongside new tasks, we are gradually refactoring the legacy code.