I don’t like waiting. And I don’t like having to verbally provide a customer service representative at my phone company or at my bank my account number after I’ve already typed it in using telephone dial pad, only to have my requested action kicked out for manual (and uninformed) intervention by a customer service representative.

I think that’s why the thought of business process management appeals to me. I like the idea that companies are streamlining their businesses processes to integrate tasks that cross systems (so I don’t have to tell them my account number all over again!) and automate tasks to enable them (and me) to accomplish a given process more quickly or efficiently.

The problem is that not every process always goes smoothly. Like the automated telephone input process that gets kicked out to a human representative when there’s a problem, exceptions are an important, and often overlooked, part of improving a company’s business processes.

In the last column, I introduced the notion of why business process exceptions are important to consider. Basically, whether we like it or not, they’re ubiquitous. For almost every process, no matter how automated, there will almost always be the need to handle exceptions. In fact, certain industries (like telecom with provisioning processes or insurance with claims processes) are rife with complex, multi-step processes that may involve multiple human-centered steps, all of which may have dozens of potential exceptions and exception points. For example, process exceptions in order management, supply chain transactions, and insurance claims processing are expensive, cause operational delays and can reduce customer satisfaction.

Let’s look a little more closely at some of the causes of business process exceptions. From my perspective, there are four main causes: system errors, data issues, external factors, and process design.

System errors can be independent of the transaction data and business logic and can be caused by underlying system problems, such as servers being down or services that aren’t available. One example of this type of exception would be when a web service times out after it doesn’t receive a response from a call to another web service. For the purposes of our discussion, we will not be focusing on these types of exceptions—since they’re driven by underlying systems issues.

Data issues are the next category of potential process exceptions and are pretty much what you’d expect—missing, invalid or inconsistent data. An example of a data issue would be an address field where the zip code is missing, contains alphabetic characters or doesn’t match the information in the customer database.