Analytics-driven approach to agile software product delivery
MetadataShow full item record
Citation (published version)Boncheva, Bogdana and Ivanov, Penko. 2020. Analytics-driven approach to agile software product delivery. Computer Science and Education in Computer Science 16th Annual International Conference CSECS 2020, Sept. 5th 2020, online, ISSN 2603-4794
Two key factors drive the software product delivery - the ideas for new products, and the latest approaches for optimized development. This paper focuses on the software development process and shows how data analytics enable innovation and efficiency in the delivery of a new product. The authors recommend the tools and techniques they have tested and proved successful in an international product organization within one of the leading media companies in the world. The presented analysis addresses the challenges of the standard practices in agile software development - continuous incremental product delivery and integration. This iterative approach implies developing and delivering features before a product, or even a product vision, are entirely complete. The method gains continuous feedback from the customer and adjusted revenue projections from the organization. The success of the approach relies on frequent and prompt decision-making by stakeholders from various backgrounds and with different skill sets. These decisions need to be well-informed as they drive rapid changes in the work prioritization and scope, and in the focus of the software development team—those frequent shifts in direction impact the delivery time and the quality of the product. Decisions on affecting the different elements of the engineering teams’ effectiveness rely on cumulative information about the teams’ capacity, lead time and throughput. This paper showcases how data analytics can drive prompt decisions and enable the necessary flexibility and improved efficiency. The authors demonstrate adapting the data visualization to the different audiences according to their interests and levels of expertise: customers, senior management, engineering teams. The paper advises how to choose the right data sets and make the correct assumptions for the data interpretation. The authors’ extensive practice shows these are the prerequisites to making the right decisions and delivering the impactful products that make an organization stand out.