The text explores the process of transforming raw data from an Apache Kafka® stream into a more structured format suitable for analytics, using managed services like Confluent Cloud and Elastic Cloud. The data, originating from the Northern Data Hub, details the occupancy of car parks in Bradford, England, and is initially in a CSV format. Techniques such as applying schemas, setting message keys for partitioning, and creating nested fields for location data are employed using ksqlDB to process and transform the data. The transformed data is then streamed to Elasticsearch for visualization in Kibana, enabling the creation of real-time analytics dashboards. The process highlights the benefits of using Kafka Connect for seamless data streaming between Kafka and Elasticsearch, and the flexibility of Kafka in creating modular and loosely coupled data pipelines. The text also discusses integrating multiple data sources, including historical data and JSON data from different formats, into a unified Kafka topic, showcasing the adaptability of the pipeline to accommodate evolving data requirements.