<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=390134628368750&amp;ev=PageView&amp;noscript=1">

Process Mining for Process Improvement

Apr 16, 2020 / by Teemu Lehto    |    15 min read

How to use Process Mining for your organization's Process Improvement and Business Process Management (BPM)?

Process Mining is a great method for continuous Process Improvement. In our most recent webinar in the Process Mining use cases series, I discussed four related topics within BPM: Deming cycle (PDCA), BPM lifecylce, process mining stakeholders, and process excellence KPIs, and how process mining relates to these. A live demo of QPR's process mining software, QPR ProcessAnalyzer, was shown to demonstrate how process mining can be used for process improvement in practice.

Process Improvement:

1. Deming cycle - PDCA

Deming cycle - PDCA is one of my favourite methods for implementing change in any organization. PDCA stands for Plan - Do - Check - Act.


In order to improve your business, you first need to analyze your business operations and identify problems that are limiting your performance. In fact, all organizations in the world have some kind of challenges either within their own operations or caused by the environment around them. From an internal perspective, one of the biggest challenges that organizations now have to face is digitalization - how to use the digital platforms to make their processes more efficient. In the current pandemic crisis, it is critical for all organizations to respond quickly to minimize risks and keep our businesses up and running.

After successfully pinpointing your organization's issues, you should identify their root causes - what actually causes these problems. Based on these analyses, you then plan the improvement. This is where process mining software proves to be a powerful tool, as it can quickly discover your as-is processes, their inefficiencies and root causes based on your own data - event logs. After implementing process mining, you'll quickly get "a complete X-ray of the processes" based on your own systems, that you already have in use. You'll gain the power to make important, necessary decisions based on this fact-based data.

deming cycle



This part happens outside of process mining. After process mining gives you an in-depth understanding of your own processes, problems, and root causes, you'll have a good idea of what to change. The outcome of the operational development change will then be visible in your transactional system data, such as the ERP, CRM, and Financial systems. 


Now is the time for you to study the results that your changes bring about, and see what is working and what is not. Process mining is able to pick up these changes in the end-to-end processes and continue monitoring them for you, in contrast to traditional analytics, which is hard-coded in and won't follow new changes in your process. Process mining helps you to verify if the changes are working as intended, and if any value is added to your process. This step acts as a bridge between the Do and the Act parts.


If the changes that you made in the Do step work successfully, you will then standardize the solution for all of your business. It's important to make sure the solution is implemented, standardized, taken into use, and that it will continue to stay in use. In other words, you should continuously monitor your processes in order to timely react to future problems. Process mining is used to monitor the whole end-to-end process making sure the process improvement works as intended and also identifies the new most important problems limiting performance.

#Tip: A good rule of thumb is to implement small-scaled changes in the Do step, then standardize or scale it up in the Act step.

#Tip: From the process management point of view, process mining is most powerfully used in the Plan, Check and Act steps.

#Tip: Remember to give process mining analysis results from the Plan step to the team implementing the change Do step, such as the RPA/Process Automation team.


2. BPM life cycle

The second process improvement topic that I'll cover is the BPM lifecycle. This model is very well aligned with the previous PDCA model, but the way process mining can be utilized with these steps is perhaps more tangible. 

This cycle starts from the process identification. No matter what business your organization is in, you have a great number of processes - the two most common end-to-end processes being the order-to-cash (OtC) process (when you are selling and delivering your products and services) and the purchase-to-pay (PtP) process (when you are buying something). From the functional perspective, your organization also has a large number of internal processes in HR, finance, logistics, warehousing, etc. You need a holistic view of all of your processes, so that you can identify which processes are key to your business, and which ones are adding real values to your customers.

As process mining works on a case level (we have a unique identifier and accurate log of events related to each case), it can show you what happens in that one case, and then in one million other cases - adding all up for you to comprehensively understand the whole process.


BPM lifecycle_v2

From a process management perspective, you first start with process discovery in order to understand what is your as-is process. It's noteworthy that the as-is process model you gain from process mining comes from fact-based data in event logs, which gives superior fact-based transparency to the actual operations for process owners and other stakeholders. These process mining models are much more accurate and detailed than traditional, manually prepared process diagrams.

Once you successfully discover your as-is process, you'll move onto the next step, process analysis, where you look for weaknesses and the impacts of them in your process. Process mining provides you with a holistic view of your process where you can easily identify the bottlenecks of your processes, the rework, as well as the exceptions. An important result from the process analysis is the root cause analysis, which is provided automatically using the process mining tool and is crucial for understanding the causes of the operational problems. 

In comparison with the deming cycle, this BPM life cycle is more complex, where your goal is to create a new process model through process redesign and process implementation. Process mining helps you to pinpoint the variations that work well in your organization. Thence, you select a combination of those variations for your new process model. Alternatively, you may implement a completely new business model and plan the to-be process based on new digital transformation possibilities. Your next step is continuous process monitoring on your newly deployed model. What process mining can do here, is to give you the conformance and performance insights, to see how your new process is performing and how it conforms to the desgined process.

The cycle keeps going on like this, increasing your business process maturity and eventually transforming your whole organization.

#Tip: This BPM life cylce can be applied to your whole organization and all processes, but you should start with the most important five to ten key processes.

#Tip: Root cause analysis often shows, that the process is working fine in some business areas but failing in others. Knowledge sharing between best performing business areas to problem areas usually helps to get rid of the problems so that there is no need to do complete redesign for the whole process.


3. Stakeholders

Now that you have fact-based, in-depth insights into your processes thanks to process mining, you should think about giving this information to the right people who will take meaningful actions to transform the way your organization works. During the 20 years that I have worked in QPR helping large organizations around the world in operatinal development, I can identify at least six stakeholders that will be heavily involved in, as well as benefited from, process mining.


BPM stakeholders


The management team holds the ultimate responsibility related to business operations, and will get considerable advantages from process mining. The CEO and COO get a fact-based "big picture" of the operations in the whole organization. The CIO is likely to be interested in how the ERP/IT systems and digital processes are working, as well as how the systems support employees to conduct their tasks and how easy it is for end customers to interact with those systems. The finance directors and the CFO can monitor the waste and rework in the organizations in order to reduce costs and increase net profits. Process mining should have at least one C-level sponsor in the management team to provide transparency and fact-based undersdtanding about business operations for decision making.  

Process owners are the key people in your organization when it comes to process mining. They manage and monitor the end-to-end process model and the as-is diagrams. In process-oriented organizations, the end-to-end process owners for order-to-cash and purchase-to-pay processes have clear responsibilities and goals related to process performance as well as resources and decision-making power for implmenting changes.  

Process analysts are those who work with process owners. They are usually on the business side, which means they analyze how the process works, what the problems are and how they can improve the situation. In other words, when process owners are responsible for the big picture of the whole process, process analysts are those who utilize the results from process mining in their work.

#Tip: Process owners own the process mining models, and process analysts use these models to improve operations in their own business areas.

Process participants are those who work within the actual business operations. Their work produces the event logs used in process mining and process modeling. In practice, the change in actual processes is a result of process participants doing their job in a different ways compared to the past. Change management deals with the potential resistance from process participants when asked to change their old ways of working. Therefore, management needs process mining in change management communication programs to make sure process participants conform to the improved processes, which is easier when the benefits and improvement results are made visible through process mining.

On the IT side, you have system engineers who implement the change into your systems, such as IT, ERP, and CRM systems. For instance, once you've seen results from process mining, sometimes you notice your ERP is not constructed in the optimal way, and that it's beneficial to make a change in the system. Other times, you discover robotic process automation (RPA) could be used to conduct some current manual tasks automatically. System engineers are those who set up these changes and make sure your processes can run smoothly. They need input from process mining (handed by process analysts) to see where the problems lie.

Last but not least, the process mining methodology experts possible located in a dedicated BPM Center of Excellence in your organization. Methodology experts help other stakeholders in using the process mining methodology and tools. They are in charge of knowledge sharing activities transferring best practices for process mining usage within the orgnization. They also communicate results and positive stories achieved with process mining as well as things to avoid in the projects. In addition to the basic process mining functionalities, the methodology experts are masters for impressive process mining functions such as advanced root cause analyses, conformance analyses, and AI-powered clustering and predictive analyses.

#Tip: While large organizations have their own BPM Center of Excellence, smaller organizations can have one methodology expert or outsource this function to a company like QPR.



4. Process Excellence

With hands-on experience of working with customers from large organizations worldwide, QPR's experts in process excellence indicate the three most important types of KPIs: happy customer, happy flow, and happy automation.

3 steps to process excellence

First of all, the happy customer KPIs means keeping your promises with customers to ensure high customer satisfaction. For example, after customers make purchases with your organization, they should receive the order on time and in full. (We have a detailed blog post about On Time In Full principle here.) These KPIs aim to achieve high customer satisfaction.

Second of all, the happy flow KPIs assure that organizations follow the agreed process in order to accomplish internal efficiency. For example, although you would manage to send the order to your customer on time and in full, but you need to use an unusual and expensive way of delivery to do this, your internal efficiency suffers. Examples of the happy flow are first time right and lead time KPIs

.Last but not least, the happy automation KPIs help your organization to monitor your process automation progress. The happy automation KPIs can applied for both individual steps and the whole end-to-end process. (Read more about QPR's solution for Robotic Process Automation here.)You can achieve process excellence for your organization when all these three steps (happy customer, happy flow and happy automation) are reached for all individual cases. This framework of process excellence can be used to improve all of the processes in your business.




5. QPR ProcessAnalyzer for Process Improvement

QPR ProcessAnalyzer has hundreds of features supporting Process Improvement. In this blog, I will highlight three features with screenshots linked to the corresponding parts of the webinar recording.

Get transparency into your processes using Benchmarking

The ability to discover the fact-based, as-is processes is at the core of process mining. QPR ProcessAnalyzer provides you with superior tools for benchmarking these as-is processes by any dimension meaningful for your organization. The picture below shows a benchmark of business in the Region: Chicago compared to the other regions. The Flowchart in left window shows that for example, 22% of cases in Chicago experience Delivery Changed activities compared to only 1% of cases outside Chicago.  The table in right window shows that while 15% of all cases belong to Chicago Region, 42% of those cases where Account Manager is Mary Wilson belong to Chicago.  


QPR ProcessAnalyzer 2020_2 - Benchmarking


Process Compliance Analysis using Conformance Analysis

Conformance Analysis is a powerful process mining feature that analyses the actual as-is process behavior with the intended design process flowchart. In the picture below, 45.2% of all cases are compatible with the Design Model and 54.8% are nonconforming. The Conformance Trend shows the monthly conformance level following with Top Violations and Top Variations.  


QPR ProcessAnalyzer 2020_2 - Conformance Analysis


Continuous Monitoring with Dashboards

QPR ProcessAnalyzer contains full Business Intelligence, reporting and analysis features, which help you communicate process mining results to all process mining stakeholders. KPI charts can be selected from ready-made presets and easily customized by any user. Moreover, detailed custom KPIs are easy to create with the powerful expression language, using the underlying process mining model. All KPIs and charts can then be freely added to custom dashboards, like the one below, showing the On Time Delivery Trend, Invoice Automation Rate and Sales Order Line Item trend.  


QPR ProcessAnalyzer 2020_2 - Dashboards

It’s a good time to take a look at Process Mining if your company hasn’t already. The capabilities and usability of Process Mining software are improving rapidly, and the market is quickly becoming mature, although there’s still much work to be done. If you think your company is ready to step it up with the future of as-is process modeling and process efficiency maximization, the fastest way to get things moving is to follow these steps:

Topics: Process Mining, QPR ProcessAnalyzer, Best Practices, Process Improvement

Teemu Lehto
Written by Teemu Lehto

Process Mining evangelist active in marketing, sales, consulting, product development and research. Teemu has been involved in 200+ end customer process mining project from order-to-cash, purchase-to-pay, plant maintenance, auditing and service. Teemu is also an active speaker delivering the process mining message as well as writer for several process mining and machine learning scientific articles. Book a meeting with Teemu using the link: https://outlook.office365.com/owa/calendar/TeemuLehtoQPR@QPR.onmicrosoft.com/bookings/


Find out more about popular Process Mining use cases through these guides:

4 Steps of RPA with Process Mining

Process KPI Reporting

Featured Success Story from EY UK:

Internal Audit and Risk Management