I've had the same experience lately. Coordinating a number of staging environments for an evolving SOA backend has been challenging for both developers and ops. In addition to the services, each one can have many other things to worry about: monitoring, error collection, which version is deployed to which environment, replicating the complexity locally to work on it...