The name of nineteenth century English mathematician George Boole may not ring a bell when you’re thinking about workflows, but in fact his work on logic created a foundation for much of the digital economy of today. Boolean algebra is all about logic, starting with a basic concept of “if this, then that.”
Boole went much further, combining conditions (“if this AND this, then …”) and excluding them (“if this AND NOT that, then …”), and Boolean algebra can become quite intricate and complex, but it follows simple rules and anyone who has ever written a computer program has used them.
Logic and decisions have always been at the root of even the simplest workflows: “If this step is completed, proceed to that step.” Automated workflows, too, can be complex, and decision points and rules for making those decisions are at the heart of how the steps in automated workflows proceed.
Workflows Are Made Up of Discrete Steps
Workflows, like computer programs, consist of a collection of discrete steps. The completion of one step leads to some other step until the entire process is completed. Some workflows are no more complicated than this, and the only “decision points” in the workflow are answering the question of whether the previous step has been completed before pressing ahead.
But sometimes, the decision point is more complicated. If the step is completed and the result is X, then a particular step comes next, but if the step is completed and the result is Y, then a different particular step comes next. Getting these decision points right is essential to making an automated rule-based workflow work.
Completion of a Step May Lead to a Choice of Next Steps
Suppose your company has an order fulfillment workflow. Maybe if the completed order weighs less than a defined threshold, or takes up a volume of less than a defined threshold, then the workflow sends it to one packing line. But if the order is bigger than one of those thresholds, the workflow sends it to another packing line. It sounds simple, but it’s easy to see how complications can arise.
Suppose a package’s weight is exactly that of the threshold weight that determines the next step. What then? Well, you could then gather further information. If the volume of the package is below the threshold volume for the decision point, it could go to one packing line, but if the volume is above the threshold, it would go to the other packing line. In other words, decisions that may appear straightforward at first may in fact come with exceptions, and your workflow must know what to do when these exceptions occur.
Making Rules for Complex Decision Points
Automated workflows depend on consistency of decisions based on any input the process may receive, and this means that you have to test workflows thoroughly before putting them into general operation. For example, if an employee starts a process to be reimbursed for business travel, how do you determine how much of the per diem allowance they get on days they are in transit?
You could have the employee enter the time they left on departure day, and the time they arrived back on their return date. If departure time is before a particular time of day, they may get the full per diem for that travel day. If it’s after that particular time, they may get half the per diem for that day. When your workflows have to do things like prorate reimbursement, you have to establish your workflow rules first, and test out as many possibilities as you can think of to ensure that the rules always lead the workflow to the correct next step.
Workflow Software Should Handle Decision Points without Human Input if Possible
Ideally, your workflows should require the minimum of human interference from the outside. You can ensure this by running multiple test cases in an attempt to make the workflow malfunction, and then if malfunctions occur, you can address those cases individually.
PerfectApps is workflow software that lets workflow users create customized processes that make perfect sense for their needs. Adding steps, decision points, and rules is easy with the PerfectApps drag-and-drop interface. Testing and modifying workflows where necessary is easy too. We invite you to watch our demo video and see how workflows based on discrete steps, logic, and rules can transform just about any work process and make it faster and more accurate.