Capillary uses machine learning to boost customer lifetime value with personalized marketing

Overview

Capillary Technologies, a loyalty and marketing platform serving brands like Shell and Levi's across 400+ brands and 450 million customers, built an ML workflow with MLflow on the Databricks Data + AI Platform to power personalized, data-driven marketing and loyalty campaigns from over 13 billion customer transaction records. Using Capillary's targeting algorithm on Databricks, campaign conversion rates increased by 600%, and the ability to target and retain customers improved from 1-5% to 200-400%, while compute and infrastructure costs dropped 35%.

The challenge

With over 13 billion customer records to analyze, Capillary's legacy infrastructure buckled under the weight of their clients' data sets, slowing performance and creating operational overhead.

The solution

Capillary adopted the Databricks Data + AI Platform lakehouse to process and analyze all client data quickly and cost-effectively, streamlining cluster management and building performant data pipelines, and used MLflow to build an ML workflow powering its personalized, data-driven loyalty and marketing campaign targeting algorithm.

Reported business value

Using Capillary's algorithm on Databricks, marketing campaign conversion rates increased 600%, the ability to target and retain customers improved from 1%-5% to 200%-400%, inference queries dropped from 150 to 115 seconds cutting compute and infrastructure costs by 35%, and MLflow saved 180 hours of MLOps overhead.