Avoiding the System Migration “Blues”

Making any changes to a business-critical enterprise system requires effective and detailed planning and design – nothing can be left to chance. This is especially true when considering a “rip and replace” migration to a new core system infrastructure. Whether it is Linux or Windows, or some variation thereof, developing detailed plans is crucial to success. However, even with the best planning, one cannot accurately anticipate the “pain points” of a new system implementation and the extent of the post implementation learning curve. Having just completed such a migration, here are some lessons to ease the system migration “pain points”.

Planning: Migrating to a new system requires significant planning, and there are number of planning aspects that must be considered. This type of detailed planning is often more time consuming and costly than anticipated.

• Requirements analysis – Requirements analysis aims at identifying the necessary data and information required for the business and enterprise. This often involves initial analysis, management expectations, key user requirements elicitation, and knowledge worker needs.
• System design – Just as the requirements analysis answers the “why”, system design answers the “how”.

                             How will the new system integrate with the current environment?
                             How will the users be affected?
                                                                       How will the new system look/act/feel?

• Implementation planning – Once the “why” and “how” questions have be answered, we need the “what”. The implementation plan provides the comprehensive technical and procedural plan for executing the migration, and provides the guide for deploying a scalable, reliable, and dependable enterprise solution. This is where the “rubber meets the road”, and this plan will detail what will happen, and in what sequence. Additionally, if the migration is a wholesale replacement, there are certain infrastructure components that will not be affected and will need to be considered and integrated.
  

Security and/or regulatory compliance is also a very important consideration throughout this planning process.

Project Management: Of course this is a vital area in that organizational management and stakeholder buy-in is fundamental for project success. These participants often determine the budgetary constraints and schedule, which can be in conflict with the project team and the realities of the environment.

Training: There are often disconnects between training and organizational reality. Even the best administrative and end-user training cannot always fill the gaps between how a particular environment/application works in theory and how it works in reality. Typically, no amount of administrative training will accurately reflect how things work with the idiosyncrasies of integration points and end-user needs. This is also true for end-users who have become accustomed and comfortable with the legacy system. Not only will they be resistant to change, they will have developed habits, good and bad, that simply will not work with the new system. This can be frustrating and aggravating to the users who often lash out at the “change agents” or management. Setting realistic expectations for end-users should be considered critical to overall success.

One of the more successful ways to facilitate easing into the new system is to implement a parallel system where users can get hands-on with the new environment. Often we’ll find “product champions” (the politically correct way of saying “guinea pigs”) taking time to learn the new environment and helping to support their coworkers, minimizing the learning curve and the time away from productive work.

Troubleshooting Methodology: As with training, there is no easy transition for determining troubleshooting methods for the new system. While basic troubleshooting logic (i.e. OSI model) is the best place to start, certain aspects of the new system, like management consoles, layers of options, and new security implications, can often fog the troubleshooting process.

While this list is by no means extensive, it is a real-life key to successful system migration. As we have done many times in the past, C/D/H can be a pivotal collaborator in the process, alleviating the project burdens from the day-to-day duties of the IT staff. We have years of experience with project management, comprehensive planning, and the particulars of successful implementation. When considering any such migration or change to your enterprise information system, C/D/H can be your partner in success.