As organizations look for opportunities to put data analytics to work, DevOps is one area that deserves attention. In many organizations, data scientists and software developers are still siloed, but bringing their talents together makes sense. Development informed by data can help teams in a number of ways.
The goal of DevOps is to shorten the time required to get ideas into practice by accelerating the development process in a quality-controlled way. Its methodology depends on a steady, reliable source of clean, organized and indexed data. Generating the right data during the DevOps process, and then using the resulting insights to inform the next iteration, can minimize error and increase the likelihood of success.
Analytics is a natural complement to the DevOps feedback loop, providing valuable insights that inform continuous improvement in decision-making. Collaboration, however, is essential. When developers know what data they need to collect, they can create code that will write logs to a file for analysts’ review. That requires not only joint planning but also clarity about what the goals are — both for development and for the business as a whole.