What is a process?
Before we get into Process Mining, let's talk Process.
What is a process? It’s not something we think about often and maybe only when we see a bad one. This might be something like a delayed flight, waiting forever to get a passport, or even sitting in an emergency room waiting to see a doctor. While we tend to jump to thinking about the human element of these situations, we rarely think about them as a process.
Looking at this from a more organizational standpoint, we might notice our bad processes when there is a delayed delivery in our order to pay process, or if there is a maverick purchase in our purchase to pay process, or even something as simple as slow time to invoice in our accounts receivable process.
Basically, a process is the way something gets done. It is a series of steps that are taken in a specific order to achieve a goal.
What is mining?
When we think mining, we think gold, diamonds, pick axes, tunnels, etc. Mining is 'the process of extracting useful materials from the earth.'
What is process mining?
If we combine the definition of a process and the definition of mining, it is extracting useful material or data about the steps that are taken to reach organizational objectives. Process mining gives us visibility into the processes that take place within an organization.
We can think about mining as an x-ray of a process. We get exposed to the innerworkings of the process - how it is completed, who is completing it, how long is it taking on average to achieve the goal and so many other KPIs. Once we have a good look at the x-ray, we can take measures to heal the process. Why? Because there is always room for improvement.
No matter how good or bad a process, the way organizations do things are always subject to change, whether that is formally to stay compliant with new best practices or informally because some great problem solver has come up with a better way to accomplish the goal.
Why do we process mine?
Processes are completed within multiple computer apps, by multiple employees in many different departments. To find out if those complicated processes are as good as they should be we process mine. Process mining is a fantastic way to understand these simple or complicated processes, find bottlenecks and inefficiencies to ultimately take measures to remove any gaps.
Process mining is objective because it is data driven, not assumption based. It is faster and more accurate as it eliminates the need for manual process mapping and it avoids the rip-and-replace method because it works with your existing systems.
Power Automate Process Mining
If you didn't know, there is now process mining capability inside the Power Automate. It works perfectly with the existing Microsoft platform and ecosystem to provide end-to-end solutions. It can visualize the processes happening in your organizations allowing key users to monitor KPIs, compare the 'As is' process with the 'To be' process, and determine root cause analysis.
It will help make faster informed business decisions and improve overall operational efficiency.
How to Process Mine?
I talked so much about the what and why of Process Mining, but the real question you may have is 'How?' How does it really work?
Well, to get started you need an Events Log for your process. At the very least, we need three important pieces of information for each single event in the log:
Case ID: Who is involved?
Activity: What is happening?
Timestamp: When did it happen?
Once the events log is ingested and proper data transformation is completed, a process map is generated. A process map shows the flow of activities for a process from start to end. Right away, you will see a map that looks like a plate of spaghetti (unless your processes are being completed perfectly). You will see some KPIs such as:
Average case duration: How long is it taking for a case to complete on average.
Self loop cases: % of cases that have an activity followed by the same activity i.e. a case went through the approval process after being approved once.
Loop cases: % a case going through a series of activities and having at least one activity repeated.
Rework cases: % of cases self looping or looping
No of variants: How many different ways was the process completed (My favorite way of explaining a variant are the following examples:
Variant 1: Wake up, brush my teeth, get dressed, eat breakfast, get to work.
Variant 2: Wake up, get dressed, brush my teeth, get to work, eat breakfast)
No of cases: Total no of cases we are working with.
No of activities: Total no of activities (Creation, Review, Approved, Rejected, Payment)
There are far more capabilities within the Power Automate Process Mining Desktop App.
You will see the performance of the process. How long is the process taking to complete in total, or on average. What has been the maximum duration, vs the minimum duration. Also, the total time it takes to go from one activity to the next.
You will see rework data such as self loop count, loop count, rework count, etc. This will help you see how many times an activity is followed by the same activity. Or how much time is spent correcting a mistake. Or how many times a case has to go back to a previous step due to missing information.
You can see how much money is going through each activity and the amount of money going from one activity to another. This information is found if you have the right data in your events log.
Within the Power Automate Process Mining desktop, you also have the ability to create different views. Say you are mining your accounts receivable process and you find out your customers are paying late. Immediately you can create the following views:
On time payments
Late payments
You do a process compare to compare your two views to find where the differences lie.
Then you can conduct a root cause analysis to identify the high-risk customers based on their likelihood to pay. The potentially troublesome customers can be identified and escalated before the late payments can get out of hand. Root cause analysis helps you find the connections to your data and get to the bottom of your problems.
You can also watch the animation play over your process map which helps to visualize the process development.
And statistics shows information about your cases and activities through visuals and charts.
Lastly, you can choose to export your data if you would like.
What happens after process mining?
So you have completed the process of process mining. You have your X-rays and you can see the inner workings of how things within your organization are actually getting done! You know how long things are taking, where the mistakes are happening, if multiple approvals are getting in the way, and a whole host of other process level problems. Now begins the hard work of making the informed decisions and effecting change within your organization to make those systems better and more efficient. Here are some places to start to make your process better:
Standardize: Make the process repeatable. It will ensure that processes follows happy path or the 'To be' path you envisioned when the process was put in place.
Streamline: Remove unnecessary or redundant steps
Optimize: Improve the quality of the process and make it effective and efficient
Automate: Eliminate the need for human efforts where applicable.
Good luck and happy process mining!
About Me
I'm Leoza Kabir Barker, a Functional Consultant at XRM Vision with a focus on the Power Platform. Through my expertise, I aim to streamline processes, optimize operations, and maximize productivity.
Connect with Me
Blog: www.leozakabirbarker.com
That's fascinating, thanks for sharing what process mining is and how it happens with Power Automate!
Fantastic post, I love the detail. Process mining is a massive growth area.