Cut-Over into Operations

Abstract

Cut-over here refers to the transition of a project into operations. Not much has been written about this very critical phase, which usually causes a lot of disruption to the business. It is a difficult phase because it is when normally two distinct and disjoint teams have to work together in transferring an asset from one to the other (project into operations). The first and foremost premise for a smooth cut-over is to involve operations in aspects which they will later on be in charge of maintaining: technical architecture, functional design, and its maintenance, as well as tasks in which they themselves will be involved in during the project itself such as developing interfaces or data migration. It is also important for operations to have a full understanding on the grand vision of the application itself early on, and to understand the details (process, code, and permissions) as the go-live date nears. From the point of view of the project team, several aspects particular to cut-over need to be addressed. First, is the awareness of the users to the upcoming go-live, which may significantly affect the manner in which they work. Other important aspects that need to be coordinated early on with Operations are their particular guidelines, the environments that will be used (DEV,QA) and who will be responsible for releases in/from each, backup and restore procedures, security, as well as escalation procedures. Data migration and data quality are interrelated and depending on the project, may carry a heavy amount of work that need not be belittled, and should be thought of early during the project. Data quality cannot be usually decided upon by neither the project nor the operations team and is the realm of the end users; it is only they that have the capability to validate the data, and therefore, is their responsibility to clean it, which should be done in preparation for the data migration and go-live. A related aspect is that upon going live, some data may not or cannot be migrated and should be manually recreated into the new system, this again requires end-user participation and should be managed early during the project. Lastly, success of the project upon going live depends greatly as well on the ability to resolve incidences as they are being reported by the end users. Crucial to this success is a clear delineation of who is to conduct 1st, 2nd, and even 3rd level support, as well as the escalation procedure process.