The introduction of Puppet on an infrastructure is a long and intriguing voyage without return. It requires planning, patience, method, experience, and skills, but it brings huge results.

There's definitely not a unique way to face it. In this chapter, we have exposed the general scenarios we might face, the possible alternative approaches, and have suggested a step-by-step procedure articulated in different phases: information gathering, priority setting, decision making, code development, application to production, and testing.

These phases should be reiterated at each step, with more or less emphasis on what matters to get things done.

We have also faced the changes that such a process involves: how we need a new mindset and new processes to sustain a Puppet setup.

Complementary to this is an effective management of our code: how it's versioned, reviewed, tested, and delivered to production. These are some of the topics we are going to discuss in the next chapter.