My last post made the case for investing in planning for large and complex changes in networks. But there are plenty of times when plans aren’t easy to execute because there are things we don’t know. Times when, for example, we’re changing a network in a way that hasn’t been done before. Or when the changes are complex and we can’t be sure what will happen afterwards. Or when the timeline defies estimation because the deployment is very large.
Pilots or dry-runs are our best friends in situations like these. Pilots are activities executed in a small and controlled environment, and they can help us discover the unknown. They’re best applied when the unknown variables are the timeline and the team’s performance. Dry-runs are simulations usually executed in a lab environment. They simulate planned changes in the lab and reveal their impact.
Now that the team has the plan, and the ambiguities have been removed by a pilot or dry-run, it’s time to execute. The plan will make the team more confident and keep everyone in sync. A good plan will mean everyone will know what to do and what to expect at each stage. But you still have to follow the plan! I’ve seen many engineers ignore a plan in favour of working in a way that they believed was more correct. In some cases, they thought that, because they had memorised the plan, they don’t need to follow it. Two types of engineers are vulnerable to this: The ones that don’t like to follow procedures and the ones that don’t know what is to follow a procedure. Ignoring the plan during the deployment stage can easily lead to skipping important steps and undermining the success of the project. The moral: If you’ve got a plan, stick with it.