Real-Time Data Driving M&S Marketing Success
M&S Baby Club Success: Real-Time Data, Scalable Marketing and Customer Growth
What is Lorem Ipsum?
Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged.
Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged.
Introduction
Marks and Spencer (M&S), an international retailer based in London, since 1884 specialises in retailing a wide range of products, including clothes, home products, food, and cosmetics.
In 2023 the UK baby clothing market was valued at £900 million, with M&S holding a 6% market share in children's wear, ages 0-2. They aimed to grow their market share and attract more family shoppers with the launch of the first of its kind ‘Baby Club’ for Sparks members.
M&S prepared for launch by creating a new minimum viable product (MVP) membership model called ‘Baby Club,’ aimed at expectant parents or those with children up to two years old. Their goal - to offer a tailored membership program providing discounts and personalised services as part of their broader loyalty strategy.
A significant challenge in this initiative was collecting the latest customer data which was spread across multiple systems. Once collected it needed to be transformed ready for reporting, all while adhering to strict data governance and privacy requirements.
This use case represents a small section of a much larger project we completed with M&S.
The ‘Baby Club’ has contributed to double digit sales growth in the category for 2023/24 and M&S now sells over 1.1 million baby grows each year.
Challenges
Accessing Source System Data
The challenge was to access data from membership databases outside of M&S’s data platform (BEAM). Data present in the source system needed to be pulled into BEAM via a daily batch process. This process presented significant challenges in data processing time, meaning customer interaction data was not readily available for analysis.
This hindered the delivery of up-to-date reports, slowing decision-making and reducing opportunities for customer engagement.
Data Management
One of the major challenges was fetching and managing historical data for membership users. In scenarios where a user opts into membership and remains signed in for an extended period, tracking their activity was relatively straightforward. However, managing cases where a user opted in and out multiple times within a short period proved to be far more complex.
Accurately recording these frequent changes became a significant issue, complicating the capture of latest subscriber action, i.e. opt-ins and opt-outs. It was crucial to preserve all historical records of these actions to facilitate detailed analysis and support the updated marketing campaign.
Data Governance and Security
Adhering to data governance and security policies presented another significant challenge. Ensuring compliance with data privacy standards was critical, especially when handling sensitive customer and child information. M&S required deletion of all past records and personally identifiable information (PII) to be compliant, however, the non- PII data, still needed to be available for analytics. This challenge needed an alternative solution.
Solution Delivery
Kafka for Data Ingestion
Kafka was employed for streaming data ingestion to facilitate the fast processing of customer data. The implementation of Kafka bridged the gap between the source system and the target Data Platform - BEAM. This enabled the system to handle large volumes of data efficiently with seamless integration to other systems.
Azure Ecosystem
The solution leveraged Azure services, including Blob storage for low-cost, unstructured data, ADF for ETL, facilitating data movement between the components. Azure Databricks for high performance transformation and Synapse Server for analytics. This combination provided a robust infrastructure for handling the ‘Baby Club's’ data needs.
Airflow for Orchestration
Airflow was used for workflow orchestration, automating the process of data movement and ensuring timely data processing.
Historical Data Preservation
The system used slow changing dimension SCD-Type 2 data model to track the historical customer opt-in and opt-out attributes, enabling detailed analysis and targeted marketing. Unlike the SCD-Type 1 model, which only captures the latest interaction, the SCD-Type 2 model stores every interaction, ensuring comprehensive data is available for analytics.
Data Governance & Security
Data Masking
A key aspect of the solution was the implementation of data masking to anonymise customer details.
To adhere to data governance and privacy policies, Cloudaeon’s data team opted to mask restricted data instead of deleting personal information or entire data layers. By choosing this approach, they ensured compliance whilst retaining the necessary data structure.
System Design
Multi Layered Data Architecture
The system was designed with three layers: ingestion, conform, and analytics. The ingestion layer handled initial data capture, the conform layer validated and transformed data, and the analytics layer provided the final dataset for reporting to the business users.
Results
Successful Deployment
The ‘Baby Club’ MVP was deployed successfully within three months, meeting all the project objectives. The client achieved near 100% accuracy data, which significantly improved their decision-making process and campaign performance.
Performance Metrics
The system demonstrated scalability and resilience handling between 6 and 10 thousand membership transactions per day.
Test Case Success
All test cases were executed successfully, with the system functioning as desired. The data masking process, in particular, was highlighted for its effectiveness in ensuring data compliance.
Customer Engagement
The implementation helped the M&S gain deeper insights into customer behaviour, enabling them to run targeted campaigns and enhance overall customer satisfaction.
Future Scalability
The system was built with future growth in mind, positioning it for easy expansion and additional features as the ‘Baby Club’ program scales.
“Sharry Cramond, Director of Loyalty, Food Marketing and Masterbrand, said: “The M&S baby club is the first of our special interest clubs for Sparks members, and we have big ambitions to make it the very best baby club in the country by bringing together the very best of M&S – from food, to clothing, to our cafes – and be a one stop shop for the UK’s incredible community of parents.”
