Personalized Commerce at Scale for a Leading Baltic Beauty Retailer

Client

A regional health and beauty retailer serving Latvian and Lithuanian markets.

Need

The client wanted to make the online shopping experience feel more personalized and relevant for beauty customers, while keeping it easy for internal management. To support that goal, product discovery had to work across search behavior, category navigation, recommendation logic, product data exports, and promotion behavior once customers moved from storefront browsing into the cart.

Solution

With proven ecommerce web development expertise, Expert Soft worked on the practical layers that make personalization visible to shoppers. The goal was to make recommendations and merchandising easier to control inside SAP Commerce and Spartacus, instead of treating personalization as a separate front-end add-on.

Centralized product data for personalized channels

We developed a unified product export system, the General Product Feed, to make product information easier to distribute across storefront and external channels. Instead of maintaining separate export services for each business unit or third-party platform, the feed supported multiple configurations from one shared foundation.

Groovy-based configuration allowed business-unit-specific adjustments without duplicating service development. This made product visibility more consistent across channels and simplified maintenance for teams working with localized catalog and campaign requirements.

Business-user-controlled content updates

For local campaigns to feel relevant on the storefront, business teams needed a faster way to adjust what shoppers saw without waiting for every content change to move through a development cycle. Expert Soft supported a workflow where users could update promotional content, product details, and localized banners through structured submissions that the platform then parsed and processed automatically.

For a retailer running local campaigns, this was not only a content-management convenience. It gave market teams a more direct way to keep customer-facing product and promotional information current while reducing release overhead for routine updates.

AI keyword recommendation engine

The client wanted to move product discovery beyond static categories and exact search terms. For this, we integrated an AI keyword recommendation feature that turned category-level intent into clickable keyword tags, each leading shoppers to dynamic pages with relevant product selections.

This helped product discovery move beyond static navigation. Customers could follow intent-based keyword paths, while the platform used AI filtering logic to assemble product selections that better matched the browsing context.

AI Top 10 product suggestions

Another discovery need appeared in high-intent storefront moments, where the retailer wanted to surface products that already showed strong business signals. So, we integrated AI Top 10 product suggestions, an external AI-based service that returned bestsellers, most viewed products, and products that frequently appeared in searches.

The back-end integration used an existing SAP Commerce proxy controller. We configured the required endpoints and API keys so the ecommerce platform could request the product lists and make them available to the front-end.

The front-end then used this data for homepage carousels and suggested products in the search box. In the broader personalization story, this gave the retailer another way to surface high-intent products without separating AI discovery from the existing commerce architecture.

Virtual try-on support

For beauty shoppers, personalization also meant seeing whether a product could fit them before they moved deeper into the purchase path. For this, we integrated Modiface – a virtual try-on experience for beauty products.

Together, these enhancements supported scalable operations across content, catalog, and campaign workflows, empowering both technical teams and non-technical stakeholders to manage and optimize ecommerce experiences more efficiently.

Challenges

Preserving promotional logic in personalized flows

Personalized discovery and campaign flows can guide customers toward promoted products, but the cart still has to distinguish between products chosen by the customer and gifts triggered by promotion rules. This became important in a Gift with Purchase campaign where the same gift items were also available for direct sale.

If customers manually added those products to the cart before meeting the promotion criteria, the third-party promotion engine could misclassify the item as a triggered gift and apply a 100% discount. Expert Soft investigated the evaluation flow and worked with technical and business teams to refine the discount logic.

The final behavior added gifts only when the criteria were met and protected manually selected products from being treated as free promotional items. This kept the shopping flow reliable while preserving promotional pricing integrity.

Business-controlled search and merchandising

The default Apache Solr integration in SAP Commerce Cloud didn’t give business users enough campaign-level control over product ranking. During campaigns, local merchandising teams needed a structured way to elevate priority SKUs without relying on unplanned development work.

Expert Soft customized Solr logic so teams could apply manual product prioritization while keeping search behavior controlled. This supported product discoverability during campaigns and made merchandising decisions easier to manage inside the storefront experience.

Technologies

SAP Commerce Cloud, Spartacus, Apache Solr, Groovy, Modiface

Conclusion

By connecting product data, AI-supported discovery, virtual try-on support, merchandising controls, and promotion logic inside SAP Commerce and Spartacus, Expert Soft helped the retailer build a more manageable foundation for relevant beauty shopping experiences.