Case Study: Retail

Luxury retail company re-engages churning VIP customers and decreases discount sales with Cerebra decision intelligence

Highlights:

International fashion retail company needed an integrated churn and demand forecasting model that accounts for variance of purchasing patterns between customers.

icon multiply

Cerebra recovered >96,000 high-risk and churned customers with total value of +$300M / year.

Cerebra successfully predicted up and cross-sells with reduced discounts for 20,000 VIP customers yielding +$80M / year.

Cerebra takes the heavy lifting of modeling every customer’s individual demand into predictions, aggregating, and standardizing the results into actionable takeaways.

About:

This Cerebra client is an international luxury fashion company. It produces and retails a boutique variety of designer clothing, textile and leather apparel, along with wedding dresses, home decor, and confectionery. The company appeals to both luxury and casual consumers with more than 10 brands in 160 stores across Europe and the Middle East.

Opportunity:

The company has a strong product design and development team with merchandisers, designers, and stylists working hand in hand with data scientists and supply chain managers. After nearly 100 years of success in adapting to customer demands in an ever-changing customer base, the company was looking to improve their understanding of a large customer base where the most value-generating customers are churning, and rates of discounts are increasing. As every customer behaves very differently with their expansion into e-commerce, brick and mortar, and other digital channels such as Instagram across 10 brands internationally, the company’s old models of hard cutoffs for communication outreach and merchandiser’s initiatives failed to address the problem.

In order to drive sales in a competitive landscape, the company adopted many discount and loyalty programs over the last 5 years. These programs have shifted the expectations of the customer base to wait for discount seasons. Moreover, application of global, blanket discounts have brought down sales revenue for products. The company’s Chief Digital Officer sought out a solution to address the discrepancy of high churn and simultaneous loss of revenue with high markdowns.

One of the traditional large consulting firms previously delivered a snapshot that outlines current consumer segments. However, they could not provide a dynamic software solution that can continuously adapt to changing behaviors and new product launches that synthesizes data from many disparate sources.  In order to address the need to improve pricing strategy by integrating behavioral data across multiple CRMs, competitor sources, text-based reviews and past sales for 10 brands, the CDO approached Cerebra.

Solution:

Cerebra solves the “Why Them Now?” problem with its proprietary Decision Engine. Cerebra automatically extracts the segments that are most reactive to promos and identifies personalized offers for every buyer segment, targeting at the right time to maximize engagement and minimize losses. In order to do so, Cerebra ingests past sales data with inventory and customer behavioral metadata to identify patterns and anomalies in consumer behavior in real-time. The cohorts with the highest potential are surfaced to the decision makers on a daily basis, ranked by urgency and financial impact. This information is used by the company to:

  • drive sales through connecting with the customer at the moment of maximum likelihood
  • increase basket size through successful cross-sells
  • reach out before customer churns
  • only offer discounts when it truly going to make a difference in buying likelihood

This information is delivered to the company as daily directed forecasts for every store manager. The forecasts are aggregated by which options, merchandise groups and brands were likely to sell to which customer segments, so that the stores can guarantee inventory.

Cerebra’s Decision Engine ingests structured or unstructured data in any format: numeric, categorical, time series, text-based, or image-based. The system translates each stream of data into a signal, and groups the signals into customer cohorts and product groups that exhibit similar behavior in relation to each other. The behavioral forecasts for each of these groups are compared against the core KPIs of the company, such as acquiring new customers, reducing losses, improving customer satisfaction, and decreasing stale inventory. Cerebra identifies patterns and causal effects across all disparate signals in real-time, providing recommendations tailored for each role, based on long-term forecasting. The insights adapt to changing customer habits and market landscape, and automatically improve based on the results of the actions taken by the company.

Results:

Reengaged customers. Cerebra identified 96,000 high-risk and churned customers with total value of +$300M / year. On a weekly basis, Cerebra retrieves more than 1500 high-risk churn customers, and the products they are most likely to be interested in.

Reduced returns & discounts. Cerebra’s demand forecast was used successfully to generate +$80M incremental revenue for 20,000 VIP customers through the goods they are most likely to buy with the highest price they are willing to pay.

Rapid Integration. On top of the initial deployment which created recommendations from past sales and inventory data, it took Cerebra only 4 weeks to integrate Web Browsing behavior into the decision making process in order to give offers and promos to e-commerce customers in real-time.

Demand-driven replenishment. After the success of the initial forecasting project, the company asked Cerebra to extend the scope of the partnership to alert the inventory and merchandising teams for quarterly option planning. Cerebra provides about 1.7 million demand forecasts for a team of +40 product designers and merchandisers.