user interaction and business data in process discovery
Figure 1. User interaction & business data in Process Discovery (comparison)

Business Process - what is it about?

I think we all associate the business process as a certain graphic scheme. Imagination suggests a well-known flow chart with a few branches at the beginning and end. Thanks to this, however complicated, we can clearly visualize processes ... and more importantly - understand.

The process model seen in a flow chart is usually totally different in real life. Researchers and Process Managers have successfully found a way to automatically analyze actual process flow using process mining tools (such as German unicorn - Celonis). They are collecting data from event logs of ERP class systems (example from MS Dynamics below) and let us get the diagram of real process flow and compare it with the model in a few minutes.

event log example
Figure 2. MS Dynamics Event Log example

Data arranged in this way allow us to calculate the exact course or number of cases closed in the specific phase of the process, which is well illustrated by the example below.

process flow chart by celonis
Figure 3. Process Flow chart in Celonis software

This approach gives very fast results and allows from a "bird's eye" view the ability to diagnose weak points of organization workflow. Process mining allows you to regain control of the process.

Instead of creating hypotheses and spending time verifying them, we can immediately look for reasons in places we would never have thought of.

Process mining is just the first step...

The above example allows a person who knows the specifics of a given process to quickly assess whether such a flow is normal and whether it serves the business purpose. If we have doubt, we must seek answers to the next questions.

Unfortunately, the next questions cannot be answered using only business data (from ERP and similar systems). That's why some companies like Celonis or Automation Anywhere released add-ons for recording the "user path". They rely on recording the employee's way of working while performing a specific task. Their effect is simple to understand but cannot be processed on a large scale and... uncomfortable for the employee.

User interaction data

Any organization thinking about a professional approach to RPA and process management, will sooner or later have to understand and then answer the question - in what way action X is carried out by the employee at the end. The reason will not be companies are wanting to cut costs but market forces pushing to implement processes that are faster and better especially related to sales support and customer care.

The so-called user interaction data will be a complementary element of business process analysis. It will allow you to quickly catch the causes of delays or errors and counteract them in the future. The module for recording all interactions used by Lab1 enables saving the way the task / process is performed in continuous mode (24/7/365) for all employees. This activity is based on passive data, which means that it does not involve the final employee in any way (it does not require any additional information nor input).

Process Discovery systems based on user interaction data exactly know how each task is performed by employees. In case of Lab1 - one day contains about 30 thousand events per employee.
event log example
Figure 4. Lab1 user interaction data example

The collected data, properly ordered, could be the basis to understand how people work. To do this, activities must be tagged by types of tasks performed.

event log example
Figure 5. Tasks tagging

The above visualization illustrates how Lab1 system understands data. Thanks to such tagged log, we can:

  • specify the number of tasks completed
  • determine the pace of task execution
  • check the course of activities in the process / task (flow chart from employee perspective)
  • search for error causes, poor performance etc.

Examples of applications of user interaction data will be discussed in more detail in the next article from Insights cycle. We encourage you to subscribe to the newsletter!

Photo credit: Pexels

Share this article on: