Large Online Retailer Increases Revenue 8% with AI Search that Shows Shoppers What They Want

The Client

Founded in 2003, our client sells next-generation fashion merchandise through two sister brands. They built a custom, proprietary technology platform to manage nearly all aspects of the business and continues to invest in technology, data analytics, and innovative marketing and merchandising strategies. Their platform connects a deeply engaged community of millions of consumers, thousands of global fashion influencers, and more than 1,200 emerging, established, and owned brands.

As a trusted premium lifestyle brand and a go-to online source for discovery and inspiration, our client offers over 100,000 apparel and footwear styles, as well as beauty, accessories, and home products. They launched an average of more than 1,500 new styles per week in early 2024, and is growing. 


Revenue increase in 2 months

>200k items

Automated ingestion of the entire product category


New styles launched per week

The Challenge

With nearly 3 million active customers as of January 2024, even minor enhancements to the customer experience could significantly boost revenue for this technology-focused company.  One specific area of improvement identified was reducing abandoned shopping carts. The company aimed to leverage AI to achieve a 2-3% reduction in abandoned carts.

Recognizing Wursta’s expertise in AI solutions, the client partnered with us to  implement Google Cloud’s Vertex AI Search for Retail aiming to elevate their customer experience.

The Results

The strategic implementation of Vertex AI Search for Retail not only addressed the client’s initial goals but delivered a series of impressive outcomes, showcasing the power of AI to transform the customer journey and drive tangible business results. Read on to explore the specific results achieved.

Show Shoppers What They Want with AI-Powered Results

Wursta integrated Vertex AI Search for Retail to replace the client’s existing product search tool, while retaining the same user experience. The solution delivers search results and recommendations improved by AI, which increased revenue by 8% within two months of deployment.

By learning user preferences and incorporating custom user behavior, Vertex AI Search for Retail can reorder search results to promote those items a user is most likely to purchase. This enables our client to provide more accurate, relevant results, better sorting, and faster results. With a better understanding of what users are seeking, the client can now deliver an improved, seamless shopping experience and 1:1 personalization at scale.

Customizable Recommendation Models

Our client has the freedom to determine the best recommendation models and customizations to use for their site. Recommendation model types include, “Others You May Like,” “Frequently Bought Together,” and “Recommended for You,” which predicts the next product a user is most likely to engage with or purchase, based on their shopping or viewing history. For example, a user may be shown shoes to match a dress in their shopping cart.

Revenue Increased 8%, Far Exceeding Goal

In only two months, our client increased revenue by 8%, far exceeding their goal of a 2-3% reduction in abandoned shopping carts. Our client sells to consumers worldwide and achieved total international net sales of   over $200 million in the fourth quarter of 2023 alone.

Up & Running Quickly, Updating Continuously

The solution Wursta delivered was up and running in less than a few months. Wursta integrated Vertex AI Search with their Google Merchant Center catalog and managed the ETL (Extract, Transform, and Load) process needed for continuous, automated ingestion of the entire catalog of over 200K items. Responsive AI models are automatically updated daily to incorporate changing user behavior, shopping trends, and the latest product offerings.

Based on the successful deployment, our client continued to work with Wursta and expanded the experience to their sister brand, which was fully functional within one week.

Integrate Google Analytics & Customize without Data Scientists

Wursta also integrated Google Analytics in order to train AI models on the user behavior data our client gathers.

Additionally, our client can tailor AI models to optimize for various business objectives, such as Click-through rate (CTR), revenue per session, and conversion rate. They can conduct granular A/B testing by applying multiple rules without needing a team of data scientists. Customization capabilities include burying specific items or boosting to support new product launches or selling excess inventory.

Back to all Case Studies