A businessman using a pick ax to mine for a gem - process mining concept
PHOTO: Shutterstock

In the abstract, process mining is a technique used to understand a business process by analyzing what’s actually happening within a system — which can, of course, be very different from what you think is happening. In the real world, said Jakob Freund, co-founder and CEO of Camunda, “process mining means digging into event logs to identify both patterns and anomalies related to a process.”

For example, it is possible to infer information about a business process for managing sales leads by analyzing a company’s event logs in Salesforce.com, said Phil Simpson, manager of Product Marketing for Red Hat. “You can identify events corresponding to the creation of a new contact, follow up by a sales rep, delivery of a quote or the receipt of an order,” he said. “These can be put together to understand how the sales lead process actually works, as opposed to how a sales manager might think that it works.”

If one were to build an IT taxonomy, process mining would sit at the intersection between BPM and data mining, said Biju Mathew, vice president of the Industry Solutions Group at Mphasis.

Related Article: How Marketers Can Plan Data Mining With R Programming

When Is Process Mining Useful

Process mining becomes useful when you know a process exists in some form but it hasn’t been formally documented, according to Freund. It also is a fit for processes that have been documented, but it appears there is a problem with the documentation. “Process mining might reveal that multiple processes actually exist where you thought there was only one, or that there are certain types of errors occurring repeatedly of which you weren’t aware,” Freund said. Or here’s a more optimistic scenario; a user could uncover a more efficient and well-executed process that’s been put in place but never officially documented.

That is because advanced process mining tools can create a complete business process model and notation (BPMN) diagram, or in other words, a graphical representation for specifying business processes in a business process mode. 

This diagram can describe a discovered process, often by deriving analytical insights from multiple sources, Simpson said. “Such diagrams often indicate typical process paths, typical exception paths and deviations from desired paths, and may also identify paths that lead to positive or negative business outcomes,” he said.

The Difference Between BPM and Process Mining

Process mining and business process management technologies are both integral to a process improvement process or project. Process mining is used during the early stages of a project when the processes are being researched and documented, Simpson explained. Once those processes are fully understood and possibly redesigned, a BPM platform will automate and manage their execution. There is little overlap between these two spheres, except to say that some BPM platforms may have process mining capabilities — or integrate with an external process mining application — in order to gain insights into a process for further improvement.

Related Article: Data Ingestion Best Practices

Is Process Mining Becoming Obsolete?

Only if you are running business processes on a workflow automation engine, Freund said. “The workflow engine is generating complete historical process execution data that can be used for business intelligence with out-of-the-box tools.”