Customer Case Study: COOP – Databricks

Customer Case Study

The Co-operative Group (Co-op) is a British consumer co-operative of retail businesses with over 4,200 locations. It is the largest consumer co-operative in the UK and owned by more than 4.5 million active members.

Vertical Use Case

Personalization and recommendation engine – using data science to create more value for their customers, members, and community.

Technical Use Case

  • Ingest and ETL
  • Machine Learning

The Challenges

  • Vast amounts of disparate data (TBs) that made it difficult to securely share and explore the data in a holistic manner.
  • Disjointed data tools and workflows impeded collaboration and slowed data engineering and data science productivity.
  • Lack of expertise in Spark and the latest machine learning and deep learning tools and frameworks.
  • Data science team had access to a curated set of data and not the systems. This led to a limited ability to develop impactful ML models.

The Solution

Azure Databricks provided Co-op with a unified analytics platform that simplifies and accelerates ETL and empowers their data science and engineering organization to collaborate via interactive notebooks to build, train and deploy machine learning models. With Databricks at the core, they now have a modern and agile data analytics ecosystem that was unified, cloud-first, and enabled fast prototyping and iteration of ML models.

As a result, they were able to greatly reduce time-to-value across the organization and accelerate the pace of delivery for new features that drive customer engagement and loyalty.

  • Unified data engineering and data science to ensure an efficient analytics pipeline from ingest to production of ML models.
  • Democratized access to their data and Spark for large-scale processing and machine learning.
  • Collaborative workspace and interactive notebooks increased machine learning competency across and improved cross-team collaboration, allowing them to greatly accelerate model prototyping for new features.
  • Simplified infrastructure management and reduced operational costs through automated cluster management and cost management features such as autoscaling and spot instances.

Databricks has removed the barriers to accessing the data across our various businesses so everyone can explore the same data and leverage a unified platform to innovate at a massive scale.

Rob McKendrick, Head of Data at Co-op