Preparing for a Data Management Implementation

The key to a successful data management implementation (like most other things in life) is adequate preparation. Based on our eleven years of data management implementation experience, here is a list of ten key preparation items we have come to learn:

1. Determine your real business needs, source of payback/justification for a system, fully document both the needs and the ROI and choose a system based on these real business needs.

2. Assemble an evaluation team consisting of knowledgeable people from both IT and the departments that will be using the system.

3. Obtain full management support. You will need management’s support to ensure that proper resources are obtained and maintained over the course of the project (especially if any rough waters are encountered).

4. Make sure your network/server infrastructure is adequate. Users may currently be storing files on their local drives. A data management system requires the use of your network and a central server. If performance is slow, users may reject or refuse to use the system.

5. Fully document and agree on the exact scope of the project. Once a project gets underway, it is very easy and common for “scope creep” to occur thus derailing and/or delaying the project. A fully documented and agreed upon scope will help avoid this.

6. Develop a written project plan (that is broken into phases). The project plan should include projected dates, time estimates, responsible parties, etc. A “phased-in” rather than a “big bang” approach to projects has been proven time and time again to be much more effective.

7. Assemble (and maintain) an implementation team. This team may be the same or different than the evaluation team. (It is recommended that overall project leadership and responsibility remain unchanged from evaluation to implementation.) It is important to ensure that team members do not get pulled away from the team for other “more important” projects.

8. Obtain sufficient training for both system administrators and end users.

9. Perform a pilot test before doing a full implementation. Nothing is worse than trying to have a lot of users “go live” on a system and then discover that the system has bugs or does not fully meet their needs. Pilot testing will help avoid this productivity disaster.

10. Develop a plan and obtain whatever hardware and software is necessary for continually performing (and validating!) system backup’s on an ongoing basis. We have found that the error causing most data loss within companies is the failure to validate whether systems are really being backed up as planned and that the data being backed up is actually restorable and reusable.