The second phase in the database-design process involves analyzing the current database, if one exists. Depending on your organization, the database will typically be a legacy database or a paper-based database. A legacy database (also known as an inherited database) is one that has been in existence and in use for several years or more. A paper-based database, as you may already know, is a loose collection of forms, index cards, manila folders, and the like. Whatever the data base type or condition, analyzing it will yield valuable information about the way your organization is currently using and managing its data. In addition, the analysis involves reviewing the way your organization is currently collecting and presenting the data. As the database developer, you look at how your organization uses paper to collect data (via forms) and present data (via reports). If your organization uses some software application program to manage and manipulate the data in the database, you study the way it collects and presents the data on-screen. Finally, you take into account how (if at all) your organization is using its data on the Web, and you review any browser-based applications that work with the database.
Another part of the analysis involves conducting interviews with users and management to identify how they interact with the database on a daily basis. As the database developer, you ask users how they work with the database and what their information requirements are at the current time. You then interview management personnel and ask them about the information they currently receive and about their perception of the overall information requirements for the organization. These interviews are an important component of your analysis because the questions you ask (or don't ask) will have a great impact on your final database structure. You must conduct full and complete interviews if you are to design a database that truly meets your organization's information needs.
Next, you use the information you've gathered from the analysis and the interviews to compile an initial list of fields. You then refine this list by removing all calculated fields and placing them on their own list; you'll use these calculated fields later in the design process. The refined list constitutes your organization's fundamental data requirements and provides a starting point for the design of a new database. (As you know, nothing is ever truly final. Rest assured that you'll extend and refine this field list further as you develop your design.)
Once your initial field list is complete, you send it to your users and management for a brief review and possible refinement. You encourage feedback and take their suggestions for modifications into consideration. If you think the suggestions are reasonable and well supported, you make the appropriate modifications, record the list in its current state, and move on to the next phase.