There is no single best method for creating a process management system, as so much will depend on the specifics of the particular application.  In this section we describe some useful techniques and principles, and discuss some of the issues to be considered.

Any system will be designed around three pillars:

• Data structure
• Users
• Events

These three work together, and the detail of one will influence the design of another.

Data structure

The best place to start is with the data structure. The normal principles of data design apply:

• Fully normalise your data
• Keep it as simple as possible

At this stage only include the fields for the underlying data; do not include ‘support’ fields for handling state or audit. These can be added in later, but it will be much easier to construct the data design in two stages, without the additional support fields obscuring the design.

Users

Next consider the different sets of users that will be working with, or accessing the system. Configure each group as a Usergroup in the User Group Manager component

Events

Start by drawing a state diagram depicting the flow, remembering to include any loops and side-states for moderation etc. This can be done using a charting tool such as Vision, or even on paper.

Once the different states have been identified, you can add in support fields to the data structure to store the current state.

After these pillars have been put in place, the user interface can be designed, and the event actions can be configured to move the data through the states.