I very much agree with Steve that culture and behaviour has to change if you are going to get the best out of a technique or toolset.

The building of a reliable build pipeline including scripted infrastructure deployment is not a trivial task but the payback is immense. I can't remember the last time a deployment failed because of environmental differences.

The bit people have to embrace is that when a mistake is made you learn from that mistake, work out how to test for the situation that allowed the mistake to be made and implement the automated process to detect it as part of the test suite. The range of tests accumulate so that failure is truly exceptional.

The time wasted on Crucifixion or Lions blamestorming is more productively spent designing out the errors in the process.

The DevOps disciplines need broad adoption within an IT department otherwise a "DevOps" department will become a bottleneck and may jeopardise the adoption. There will always be specialists but the basic skills should be embedded in the teams so the specialists can focus on the thorny problems.