Workflows are great for standardizing business processes and documenting enterprise knowledge. Automating workflows can streamline operations and help you respond to your customers quicker than your competitor. But, what happens when existing workflows processes don’t work when new and unusual situations come up?
While many interactions may be similar and the established workflows work well, not all situations are the same. As a rule of thumb, 80% of situations or requests might be addressable via a defined workflow process. Because these situations are common we can build workflow process automation that is not overly complex to address them because the number of variables is limited. The challenge is addressing the 20% of situations that do not fit into these workflows. It is these edge cases that help differentiate market leaders.
In some cases, organizations may just expand traditional workflow processes to accommodate these situations, or they might just address them manually as a unique case. The former option can lead to workflows that become excessively complex and the latter introduces inefficiencies and increases the probability that enterprise data and knowledge will be lost.
A third option is a business rules-based approach. This is a strategy that manages complexity while leveraging the efficiency created by automated workflows. In a traditional workflow model, each potential path is documented in the workflow. This makes it less flexible and less capable of adjusting to new cases. A rules-based approach is much more effective because it enables a user to customize the workflow to fit the current situation.
When context-aware rules are embedded in a traditional workflow process, they can customize what actions and paths are most appropriate based on the current situation and present them to a user. The user can then choose the most appropriate path based on existing circumstances. By building a workflow around real-time data and constraining possible next steps instead of dictating a certain path, every single possible option does not need to be documented in the workflow process. With this approach, users can have a role in managing complexity. For example, in a visit to your doctor, the physician has some basic protocols they follow but potential actions the doctor can take or orders they can write are almost infinite. Building a rigid workflow process that incorporates all the possible actions is extremely complex and burdensome. If the workflow was able to dynamically change as new information is gathered this rat’s nest of processes could be simplified.
