Process Mining – from Definition to Implementation

August 21, 2020
Process Mining – from Definition to Implementation

An introductory guide – can you dig it?

As Gartner defines it: Process Mining is a technique designed to discover, monitor and improve real processes (i.e., not assumed processes) by extracting readily available knowledge from the event logs of information systems.

More simply, Process Mining identifies process inefficiencies by following the digital trail of data through a targeted system.

Click here to review 21 examples of how businesses recognized success through Process Mining.

The concept of Process Mining has been around for several years and has been very popular in Europe. It’s only recently become more popular in the U.S. as sophisticated digital technology solutions make it a better alternative.

How does Process Mining Work?

Process Mining uses your company’s own data to quickly identify inefficiencies, where to improve operations and how to get your desired business outcomes. Your workforce, customers, vendors, and others generate data in the form of digital footprints any time they use your internal systems to perform work. These digital footprints are recorded in ERPs, CRMs, databases, log files, audit tables, Excel documents and other systems and applications. Process Mining follows this digital trail through a system to discover what’s happening and automatically creates a visualization of the process from the data.

As Gartner states: “Process Mining helps you look at your business processes in an unbiased and transparent manner.”

[Check out our comprehensive resource on process mining: Ultimate Guide to Process Mining]

How is it better than traditional process mapping?

Traditional process mapping is a time-consuming exercise involving many subjective interviews and manual or boxed-software flowcharting, often ending in assumptions on what the problems are and how to fix them. It has no “staying power” once completed.

Process Mining is based on real data – not assumptions or opinions – and is significantly quicker and more accurate than mapping interviews and flowcharting. Because it’s based on data, clearly measurable impacts from changes can be documented. Going forward, Process Mining can become process monitoring – analyzing the targeted process for current performance.

Think of it this way: Process mapping is like asking a lot of people for directions. Process Mining is like using your GPS.

What are the benefits of Process Mining?

As stated above, Process Mining uses your own data to quickly identify inefficiencies, where to improve and how to get to your desired business outcomes. You’ll gain greater insights into how your company runs. Because it’s based on real data, a Process Mining project can guarantee the improvement to the targeted process, and monitor the process going forward. In fact, Process Mining can support your entire digital transformation strategy – it allows you to track and measure whether your digital transformation initiatives are working as planned in day-to-day operations, or if they need to be adjusted.

Where do I use Process Mining?

Process Mining can be applied at any point and level: a process, an operation, or end-to-end through an organization, and applies equally well to any business in any industry. Currently, clients have found Process Mining extremely useful in the following scenarios:

  • Robotic Process Automation (RPA) – As noted earlier, Process Mining quickly identifies inefficiencies in your processes that cost you time and money – especially those that waste your employees’ talent on tedious, mundane and/or manual tasks (like data entry). RPA can run repetitive, mundane tasks, or bridge previously manual data entry, 24 hours a day quickly with extremely low error rates. It can be scaled as needed to help with seasonal or event-driven changes in volume. You can even add elements of intelligent automation and artificial intelligence to help with decision making. RPA frees your employees up to do more important, value-added work.
  • The customer journey – Understanding your customer journey will get your entire organization pointed in the same direction, improve your customer experience and give you the insight to outperform your competition. Every day your customers are generating those digital footprints, and Process Mining will show you how those footprints move through your company – visualizing and diagraming the actual digital journey with hard data. This is insight generated by the customers themselves, not your company assumptions. Also, Process Mining continues on as process monitoring – analyzing the customer journey against the procedures or practices you put in place.
  • Today’s work-from-home – Covid-19 has created a whole new scenario: work from home (WFH) as part of the new normal for the foreseeable future. How have your processes and procedures changed – intentionally or unintentionally – to accommodate all the people now working away from the office? How do you control – or even recognize – these changes? Process Mining shows you what is actually happening based on the employee event data captured in your enterprise systems. Process Mining will diagram the actual (including variations, exceptions, gaps and siloes) against the policies and procedures you want.
  • Data governance and data privacy – Process Mining shows you how data moves through your organization. This helps you complete privacy and risk assessments, find information siloes and gaps, and assist with your enterprise data strategy. Process Mining can help validate a process audit or compliance procedures by diagraming the actual against a defined standard, best practice or instituted policy. This is extremely useful for data privacy regulation compliance like CCPA and GDPR.

How do I prepare for Process Mining?

  • Know what you want: Define clear business outcomes and key performance indicators (KPIs) targeting a specific situation, like customer experience, inefficiencies, or data governance. You’ll achieve a better return on investment. As Gartner states: “It is important to link your optimization efforts with the business outcomes you wish to achieve. Include the business outcomes that are relevant to your organization. This can help you to achieve buy-in from the leadership team at every stage. Some major key performance indicators and business outcomes to track include revenue, customers, products and employees.”
  • Ensure that your target data is clean and in good order, so its use is optimized. Your data may be stored in many different systems or data bases, and maybe it’s incomplete. You need to make sure your data can be extracted and collected. The benefits of Process Mining are directly related to the quality of the information that goes into it. At the most basic level, your Process Mining project data needs three attributes: time stamp, unique case ID, and an activity ID.
  • Have the right mindset. This may be a new way of doing things for you and your company, and you may have had poor experiences with traditional process mapping. Put that all aside and implement and leverage Process Mining without any human biases or assumptions – let the actual data tell the story.

Examples of Process Mining in action

Here are summaries of some case studies compiled through Gartner:

Consumer Goods – process inefficiencies

Process Mining project included:

  • Analyze and visualize carton movements throughout five production plants and a central cold store
  • Analyze the carton movements from in-feeding plants to central cold storage
  • Identify primary and preferred process flows
  • Identify and compare variants and associated flows
  • Identify re-loops and bottlenecks and quantify the effect on target efficiencies
  • Identify process inefficiencies and opportunities for change

Financial – Insurance Claims

The scope of the project was to analyze the claims submitted in case of water leakage in private or business buildings. The objective was to find measurable areas of improvement and anomalies inside the process. Different behaviors inside the process were isolated and different process dimensions were analyzed. A comparison between different regions was also performed. As a result, the company was able to find emerging best practices both process-wise and organization-wise. The company was able to define key performance indicators (KPIs) for measuring the process health and to set up improvement initiatives.

Financial – repetitive tasks and RPA

The company conducted a data-derived performance analysis through Process Mining to identify the costly and repetitive tasks that would be ideal for ideal for robotic process automation (RPA). Thereafter, Process Mining was used to monitor whether the optimal outcome was reached.

Financial – car accident claims

The project analyzed submitted car accident claims. The objective was to find measurable areas of improvement and anomalies inside the process. Process Mining isolated different behaviors inside the process and analyzed different process dimensions (e.g. different kinds of settlement, geographic partitioning of claims, compliance with regulatory deadlines, etc.). A comparison between different regions was also performed. As a result, the company found emerging best practices for both the process and the organization. Hot spots inside of the process were identified and addressed with specific initiatives and change requests.

Financial – customer onboarding

The scope of the project was to analyze the customer onboarding process across four countries. Strong differences in performance across these countries had been previously observed. The main goal of this project was to determine which steps or pathways of the process were determining these differences in performance. Process Mining of the entire process and variant analysis with filtering, side-by-side comparison, and multi-log animations allowed the company to find specific pathways that explained lower performance in some countries.

Financial – customer journey

The project looked at the car policy renewal customer journey. It analyzed the full process after a car insurance policy has run out and what actions lead to highest probability rates for renewal from customers, identifying factors such as individual customer groups characteristics on probability and increase in scope or size of the policy.

Financial – forex process

The scope of the project was to discover and analyze the paths of different forex exchange deals at the client. Data was extracted from different systems and aggregated into a single event log (each deal can embrace several systems). Next, different variants of a deal where segregated based on logical drivers (e.g. type of deal, broker/no broker) and performance driver (e.g. average number of activities, or average cycle time). The variants were then used as input for variant analysis in order to identify root causes for deviations.

Financial – loan approval

The project measured processing time for loan applications, resulting in the creation of a dashboard for monitoring process performance.

Financial – loan documents

Non-post and branch validation processes detected inefficiencies and potential savings – up to 60%. Funding and onboarding processes: detected inefficiencies and potential savings – up to 40%. Documents production process (HELOC processes) – list of data fields transferred between applications – time wasted on transferring data fields between applications – detected inefficiencies and potential savings – up to 33%.

Financial – process improvement

Process Mining replaced staff interviews with automated capture of the employee activities in real-time, producing the end-to-end process visualizations and measurements. Results: 88% time savings for process analysis improvement work.

Healthcare – medication and supplies tracking

There were several different departments (specialties) and each department had a dedicated pharmacy at the hospital building. Process Mining was applied to identify how medications and supplies were transferred between pharmacies, how restricted medication were handled, and what were the purchasing optimization opportunities that could drive cost reduction and patient care improvement.

Healthcare – invoice verification

Process Mining helped gain a deeper understanding of process challenges and deviations and identify opportunities for process automation. Results: Saved 900 man-hours per year through process automation; identified the cause for 97% of process deviations; revealed over 1,000 internal violations of the service level agreement; achieved higher degree of standardization

Industrial – supplier payments

The company’s main objective is to ensure that suppliers are paid within a reasonable period based on regulations, otherwise high penalties are due. The company looked to identify where payment time is lost and processes with low automation. They used Process Mining to discover process violations in accounts payables, discover non-compliant process behaviors using a conformance check, and monitor payment related issues with a defined set of key performance indicators (KPIs).

Industrial – customer journey

Process Mining obtained a clear visualization of the journey of the customer. Looking into the visualization, a new product development process was discovered. Instead of only producing a sample, in the new product development process pieces needed to be designed, produced and delivered quickly. By shifting priorities, the company was able to produce customer samples quicker without impacting the regular production lead times. This allows the company to grow their business, while keeping up the delivery performance for their existing customers.

Industrial – credit and collections

Gained a competitive advantage by automating and optimizing the process beyond the capabilities of the established process management systems. Result: saved more than 100,000 Euros and 1,600 man-hours per year on extra manual work; reduced execution times and idle periods in the process by up to 80%; identified compliance risks in over 70% of all cases

Manufacturing – maintenance and repairs

A machine recovery is a procedure of tests and calibrations to get a machine back up and running after repairs or maintenance. The ideal recovery was described by a procedure containing a sequence of 140 steps. The company used Process Mining to compare the recoveries with the procedure to identify the key deviations. In this way they were able to find steps that are not part of the expected recovery procedure and improve the process.

Manufacturing – equipment usage

The case study shows how Process Mining can be used to analyze the system usage of an MRI machine. It helps to understand how the customer (the physician) uses the MRI system, and how its behavior deviates from the expected (and designed) behavior.

Telecommunications – RPA initiatives

The company began an RPA initiative to improve service quality and customer journey and added Process Mining to their automation strategy to verify operational tasks that could be automated and rank them by ROI. Use cases include service desk, customer interactions (tickets from different sources), remote troubleshooting, field services and operators/workforce advisors.

Telecommunications – customer journey/automation

The company’s the main customer service portal – an interactive voice response (IVR) system – had more than 500,000 daily calls, with most of them being forwarded from the automated system to the employee-manned call center. Process Mining visualized the customer journey and the improvement opportunities to increase customer satisfaction with the IVR and reduce human interaction to the more complex exceptions.

Telecommunications – potential fraud

The company used Process Mining to understand the SIM card activation process and investigate internal stores and partner vendors for patterns that could indicate fraudulent behavior.

Telecommunications – debt collection

The company conducted a general investigation of the debt collection process, identifying customer payment habits and factors driving successful debt collection.

Why Prolifics for Process Mining?

  • Our extensive experience in banking and finance, healthcare, insurance, mortgage, public sector, retail and distribution allows us to bring our industry know-how to your specific process analysis.
  • You’ll see a quicker outcome – on average about three weeks – while keeping costs in check.
  • Our expertise in hyperautomation and artificial intelligence can help you solve issues that processing Mining finds.
  • You’ll have a “one-stop-shop” with our integrated offering of infrastructure, security and testing, along with our automation.
  • Because it’s based on real data, a Prolifics’ Process Mining project can guarantee the improvement to the targeted process.

Talk with us

The “New Normal” is here. You have a vision – don’t let your technology slow you down. Our Process Mining solutions and experience will get you there. Sit down with us – let’s talk about your challenges, review and reevaluate your plans and get you started where it makes the most sense. Vision to Value. Faster. It’s not just our tagline, it’s what drives us. It’s how we deliver solutions and services. It’s our commitment to you – and it’s needed today more than ever. Visit or email

For a larger conversation about Process Mining, watch Innovation Sandbox Episode # – Process Mining with Anant Gupta, Global Head of Digital Automation and Cloud Solutions with Prolifics.