DataOps: The Authoritative Edition (Book Review)

Since the dawn of professional practice, the community of software professionals has been continually improving processes and practices together. The agile movement fully embraced simplicity and learning, establishing principles that helped developers and testers more accurately capture requirements and specifications in the code, maximizing value. DevOps practices are helping to shorten time-to-value for stakeholders while simultaneously simplifying the execution of tests, builds and deployments that previously took days or weeks and making it easier to maintain software in production. While organizations still depend on software, they also depend on data. In the initial stages of maturity, data is gathered and pre-processed, and delivered to those who need it in spreadsheets, slide decks, or interactive dashboards. There is a time delay between identifying the need, having a developer or business intelligence analyst respond to that need, and being able to make business decisions based on that information. Similar to software development, this decoupled approach also means that multiple cycles of iteration may be needed before the business user can get what they need. “DataOps” is the label given to process improvement and automation geared toward rapid delivery of data and information. In the words of this book’s authors, “it creates continuous data flows with automated processes and self-service tools so that users can discover and deliver data by themselves in days or hours.” DataOps incorporates tools and techniques from agile and DevOps to make this happen. In this quick read, authors Schmidt and Basu introduce readers to DataOps, bringing a combined six decades of experience with enterprise data management. They start by grounding DataOps in the historical context, shifting to summarize the main tasks and services of a DataOps team. Next, they provide guidance for setting up and managing this team, from both technological and cultural perspectives. The bulk of the book explains DataOps practices: continuous design, continuous operations, continuous governance, continuous data, program execution, and design operations. With the exception of the latter, all of the practices have analogs in software development, and may even provide some transferable lessons. The book concludes with checklists to help you jump start your DataOps practice, and two case studies that describe how organizations used these lessons to deliver real value. As senior leaders, the authors emphasize how to communicate financial value throughout this process. Although the case studies are short, there is substantial value in this little guidebook. Even if your DevOps or DataOps are already in place, these authors provide well grounded insights and applicable lessons that can help you advance your organization’s data maturity.