If your company hasn’t gone platform yet, it can still be a big leap. You can still speed up software delivery by addressing your company’s change management process. In this chapter, we will examine the change management patterns we learned internally. We’ll show you what works, what doesn’t, and how to leverage DevOps principles to turn change management into an effective, enabled process.
Over the past decade, we’ve seen the practice of DevOps revolutionize the way software release teams work. Here are the most notable changes.
“The problem itself doesn’t change, because change happens all the time; The problem is not coping with change when it comes.” Extreme Programming: Embracing Change by Kent Beck
Even though we have seen delivery teams successfully transform their thinking and practices, it is still much more difficult to change entrenched structures and processes in a large organization. Change management is one of the most difficult processes to change.
Moving to a new way of doing things requires leadership support, organizational discipline, and a lot of collaboration and collaboration across all levels of the organization. However, the large legacy environments that grow in most large organizations are not easily broken up and redesigned. They are typically maintained by many different teams, each owning a piece of the technology stack. Teams that understand the work often lack the authority to approve their proposed changes; Instead, change approvals are often assigned to committees that are out of touch and don’t know enough.
All of these layers exist because large legacy environments are where most of an organization’s business is. So any change feels risky, and there’s a lot of process and bureaucracy that feels like it’s protecting the business.
Unfortunately, all of these processes get in the way of organizational growth. They simply can’t deliver software fast enough — either to external or internal customers — to meet business needs. At the same time, competitors that make their change management more effective can release quickly and repeatedly, putting them ahead of the pack.
DevOps evolution and change management effectiveness
We wanted to see if the effectiveness of change management correlated with the growth of DevOps. To measure the effectiveness of change management, we observe it from the following three dimensions:
** Success rate of implementation. ** We looked at change failure rates and deployment frequency. Ideally, companies should be able to change more frequently, recover quickly from failures and learn from them. ** Efficiency level. ** We want to know how efficient changes are the management process is based on: • Mandatory waiting period of less than 2 weeks • Changes only need to be approved once • Changes are implemented correctly and do not need to be reversed • Approved by people with the appropriate skills to make a correct assessment • Little time is required to document changes ** Performance sentiment. ** We developed this indicator as a proxy for an objective assessment of each respondent’s organization. We asked respondents whether they change in the company management program: • to reduce risk, reduce service event related downtime, provide useful information for organization, to ensure that with the appropriate stakeholders to share knowledge and information, to speed up the change of the business requirements, according to the evaluation of change risk level, provide the appropriate level of review and approval
These three dimensions — success rate of implementation, level of efficiency, and performance sentiment — constitute our measure of change management effectiveness.
We found that as organizations evolved their DevOps practices, the effectiveness of change management increased. It’s not a huge difference, but it’s statistically significant.
Change management approach
To investigate change management, we asked respondents about some of their different practices in the workplace. These can be divided into two parts: the change approval process and the degree to which the change is automated. It can be divided into four groups: mature operation and maintenance. High level of process and automation. Engineering drive. Attach great importance to automation. Focus on governance. High emphasis on human approval over automation. Temporary type. Not focusing on process and automation.
What drives the effectiveness of change management?
When looking at the overall effectiveness of change management, engineering-driven firms were found to have the highest level of change management effectiveness, AD hoc firms had the second highest success rate due to lack of process, and the remaining two groups relied heavily on orthodox approval and scored poorly on effectiveness.
Our data reveals a few things about the effectiveness and efficiency of impact change management:
Orthodox approval reduces efficiency; Automation gives teams confidence in change management; Granting permissions leads to greater efficiency.