Nine Tips for Transitioning to Agile Development Process

The overused buzz phrase “digital transformation” is simply another name for what the IT business has been doing its whole life: renewing IT systems. Enterprises of all types, regardless of size, industry or location, now can walk this path using Agile development and workflows to get applications to market in less time and with higher quality. This is not only to improve technical operations but also to impact the bottom line, for all those CFOs who are paying attention. In this eWEEK slide show, Pete Pickerill, co-founder and vice president of product strategy at Datical, discusses best practices for how to lock down the Agile transformation process.

While technology executives tend to focus on two of the three major application stack components (infrastructure and application code), there’s one element that many ignore: the database. The database doesn’t support the application; it is the application. If the goal of adopting Agile is to deliver innovation and a better customer experience faster, excluding the database team from the process will ensure that this goal is not realized because of the continued impact of database deployments.

Database teams see themselves as masters of the all-important data. However, they are largely overlooked, handle repetitive tasks manually and, for the most part, are trained to resist change to ensure the safety of the data platform. It’s imperative to educate database teams on the benefits Agile can bring to the application deployment process, seek their input and determine where automation tools can help the database team both in the transition to Agile and in their work generally.

If an organization is serious about transitioning to Agile, it is imperative that management take ownership and clearly and continuously message the initiative. This starts in the C-suite. Elsewhere, the messaging must define the benefits and action plan by department and/or group in such a way as to make plain that this is not change for its own sake; that it benefits all involved, including stakeholders and customers; and that issues such as risk have been duly considered as part of the process.

Change is unpleasant, and enterprises often take a risk-averse approach to business. However, when considering the transition to Agile, remember that the point is not to abandon willy-nilly things that have worked for decades. It’s to intelligently eliminate inefficiencies through collaboration. It’s not about going fast at the expense of safety.

As the word “transformation” implies, the process of moving to Agile should not be seen as one of rip and replace, but instead should be a measured one that builds on small successes and learns from discrete, containable mistakes. A best-practice approach begins with a documented view of the current process. This is reviewed, and feedback is provided with all involved to determine what works well, what needs improvement, what is core to operations and so forth. This can then form the basis for initial steps of the Agile transformation.

Many daily database processes are performed much in the same way as they have been for years—recovery and backups, installation and configuration and ETL all still require a database professional’s time and expert eye. Unfortunately, this means that this highly skilled resource spends less time on more strategic initiatives, such as performance tuning, disaster-recovery planning, capacity and continuity planning, and major upgrades. Automating repetitive tasks as part of an Agile transition can elevate database teams from being seen as function-oriented to solution-oriented. The key to getting this right is to start slow, automating simpler tasks one at a time and building on those successes.

Agile came to the fore within the software development community, so it is likely that an organization’s development teams are familiar with, if not already using, Agile methodologies. The challenge for these teams will be to see past potential frustration at those not doing things “the right way” and help others visualize the opportunity that Agile holds.

The database change process is often the biggest obstacle in terms of speed to market because there are fewer tools to support DevOps for the database than DevOps for the application. DevOps engineers must keep an open mind and continue to incorporate database update testing in the application release process early and often. It’s also important to test the entire application update (database included) and collaborate with development and database teams to uncover the potential bottlenecks.

Agile is about collaboration, which means that a successful transition is based on teamwork: each department and individual contributing equally to create a process that helps the entire organization achieve its goals more efficiently. Everyone who is a part of the application release process—executives, database pros, developers and engineers—has his or her role to play as Agile continues its rise in the application development space. Keeping the big picture in mind while taking things step by step will move the team—and the organization—forward.