Machine Learning helps a leading US retailer increase Campaign ROI by 285%

About

A leading beauty and personal care manufacturer in the US with a strong online and offline presence.

Challenges

  • Mass acquisition campaigns  resulted in poor conversion rates and high campaign spends. 
  • The quality of acquisitions were also very poor leading to customers with low average order value (AOV) and poor loyalty.
  • The marketing team  found it difficult to recuperate their CAC -resulting in very low ROI for their acquisition efforts.

Solution

As a first step, data from multiple sources like CRM, POS Data, online purchase data and Campaign Management System was integrated to create a customer 360-degree view datamart.

An in-depth exploratory analysis was conducted to explore key insights to identify the typical profile of high-value and loyal customers. Attributes like demographics, purchase behavior, browsing pattern and campaign response were considered for the analysis. 

The analysis helped the client to identify the behavior of the high-value customers who contributed to more than 67% in terms of sales revenues. 

As a next step, Tenzai helped the retailer to arrive at typical customer profiles or genomes for every demographic and purchase behavior segment. 

 Then using machine learning and ensemble techniques a lookalike or clone model was built . This model helped the retailer to identify prospects exhibiting attributes similar to  the high-value customer segment. Based on the propensity scores, the prospects were prioritized for the campaigns. 

For geotargeting, the Tenzai team collated the zip codes of all clients’ retail outlets in the US. The zip codes were used to identify prospects  living within a 10-mile radius of the outlets. Based on store proximity, the final target list was chosen for the campaigns. 

To arrive at personalized messaging, Tenzai once again employed the insights from the customer genome. For each demographic segment and microsegment,  top 2 to 3 category preferences were identified. The insights from the analysis were used to create hyper-personalized content for contextual emails and deliver personalized messages for prospects.

Why not use data to gain insights, guide efficiency efforts, and make predictions? Contact us.

Results

The integrated customer acquisition solution was deployed in the customer’s cloud environment and integrated with the existing CRM and campaign management systems.

The new acquisition model helped the clients to design campaigns and marketing strategies specific to each prospect group. Using the model the client could focus on high-value and loyal prospects thereby enhancing the overall quality of customer acquisition. It  led to an increase in conversion rate and reduced campaign costs thereby increasing the ROI for marketing spends.

Key benefits include

  •  A 153% increase in  conversion rates for acquisition campaigns post-adoption of the ML-driven customer acquisition strategy.
  • The acquisition model helped acquire high-quality customers whose average order size was 32% compared to the overall customer base. 
  • Focused targeting efforts helped the retailer to increase campaign ROI for acquisition campaigns by 287%.