Few people want to do things the hard way. If you had a choice between wandering a maze blindly, or using GPS to find your way as quickly as possible, wouldn’t you choose the latter?
Maybe you’ve heard about process mining technology, but you’re not sure what it is or how it works. Maybe you’re still evaluating how this transformative category of software fits with your industry, the size of your business, or the special challenges you face. Your obstacles may be unique, but the path to overcoming them must begin by examining what’s really going on in your business, and understanding how your work is really being done.
Find the answers you need in this paper, as well as examples from other businesses who have achieved great success using process mining. And discover how, like them, you can optimize your processes and truly transform the way your business operates.
Process mining is a category of revolutionary business technology that gathers log data from any of your existing IT systems and shapes it into a real-time, comprehensive visual of how your processes really operate
Process mining technology uses data to provide an objective overview of your operations, eliminating guesswork and enabling a common understanding to facilitate greater collaboration in process optimization, and transformation, across an organization
Process mining technology is flexible. It can serve endless use cases, including the optimization, implementation and monitoring needs related to digital transformation initiatives
Identified as the clear process mining market leader by analyst communities, Celonis has the innovation, experience and customer success learnings that your business needs to undertake any initiative to transform the way they operate.
Change is the name of the game, and the rate of change is increasing. In our everyday lives that means increased conveniences. In the business world it can be a boon, but it can also lead to new pressures and concerns.
Think about the things we see around us every day. The cash register is nearly 150 years old, and largely unchanged in function and format. But in that same amount of time we’ve moved from typewriters to tablet computers, telegraphs to mobile phones, and motor carriages to driverless cars. In a mere fraction of that time we’ve moved from fax machines to SMS, the first airplane flight to manned space stations, and introduced electronic payments and humanoid robots.
It’s not surprising, then, that accelerated change is a focus of business executives. The PWC Global CEO Survey 2018 indicates CEOs are optimistic about the global economic outlook even while harboring reservations about the growth of their own organizations. Although the survey indicates “the threats that trouble CEOs are increasingly existential,” over-regulation and increased tax burden are present in each region surveyed, with skill availability and the speed of technological change a concern in almost all regions of the world.
According to the NFIB April 2018 Small Business Optimism Index, nearly 30% of small businesses in the U.S. are planning major spends in the near future, and as qualified workers continue to be an issue, these funds are expected to go toward “training and labor-saving technology.”
This is a trend Forbes is predicting for larger firms as well. Gaps while looking for qualified candidates cost employers significant money, and more are looking to invest in the resources they currently have to meet the pressures of competing in an ever-changing marketplace.
Risk management, compliance, conserving financial resources, conserving employee resources, exploring ways to foster business agility, and adapting to the rate of change. In many cases businesses are looking at familiar problems wrapped up in new packages. And much of what’s driving those new complexities is our increased reliance on technology.
We develop and use more information systems every day, and continually expand the data capacity of those multitude of systems. The very nature of the technology we use means every step of every activity we make is logged, adding to the ever-expanding ocean of so-called “big data.” And this is without everyone on board yet.
Digital solutions are driving an unprecedented need for process improvement and process transformation across different industries and businesses as companies watch their competitors expand their use of technology to gain an edge. But implementing or upgrading to new technology can be expensive, and intimidating.
In fact, in a 2017 Deloitte global survey of CIOs, respondents identified the following
barriers to digitally transforming their businesses: employees who are resistant to change and/or lack the skills needed to implement and maintain changes appropriately; questions about the investment in such a significant undertaking and the perceived ROI; and the lack of a digital transformation strategy.
Yet according to the IDC FutureScape Worldwide Digital Transformation 2018 Predictions:
By 2019, 40% of digital transformation initiatives will depend on AI support to provide timely insights for better operating and monetization models.
By the end of 2019, spending on digital transformation will show a 42% increase from 2017, reaching $1.7 trillion worldwide.
By 2020, 60% of businesses will be in the process of implementing fully articulated organization-wide digital platform strategies.
As if internal pressures and competitor jockeying weren’t enough, expanded customer expectations are another compelling driver for the need to move to digital. Businesses must now contend with a multigenerational, tech-savvy customer base. In addition to speed, accuracy and convenience, these stakeholders expect appropriate management of big data and its usage.
Spotlights on data vulnerability and legislation such as the General Data Protection Regulation (GDPR) add a new layer to compliance concerns. Businesses already oversee adherence to existing policies, trying to avoid costly deviations like maverick buying. Many industries also have compliance concerns related to health and the environment. How to manage and monitor data on top of those existing compliance concerns can mean significant new challenges for modern businesses.
Still, in the face of so many other considerations, the question becomes not one of if you should you digitize, but when and how.
Executives of a wholesale electronic distributor asked themselves that same question. With a 25,000+ product portfolio and 24-hour delivery they appeared successful but felt pressure from clients and competitors to continue to deliver more.
In 2012 they implemented an SAP HANA system for warehouse optimization, but two years later feared a breakdown in processes might be hindering their efforts at faster service. Getting to the root of the issue, they said, was “vital for us because a wholesale company ultimately has only its processes as an optimization tool for competitiveness.”
They used an AI-powered technology to analyze their existing system, and in only two weeks they had the results–and identified an unexpected source of trouble. Yet the insights they gained allowed them to make the needed changes and improve their delivery to same-day service.
So, what was this method that helped them uncover a potential upset and optimize their service to customers, and what were the steps they took to get there?
“We cannot solve our problems with the same thinking we used when we created them.”
Businesses have used a variety of techniques to address their challenges and help them improve the way they operate, including modelling, simulation, business intelligence, and data mining. Yet these methods often involve extensive use of consultants or specialists, which cost time and money.
They are also limited in scope. Non-empirical methods focus on process optimizations may be hampered by assumptions about how things are working and what needs to be examined. In addition, interviews conducted as part of these efforts are subjective, and may lack details about everyday tasks or focus disproportionately on the positive or negative aspects of the process in question.
A data-centric approach may help avoid the politics of those who act (or don’t act) due to internal power struggles or lack of insight into issues, but data mining examines findings for a particular issue and doesn’t take a holistic look at the surrounding process.
There is an approach, however, that can show opportunities for improvement at the task level as well as overall process flow and deviations. It can also identify root causes of issues and offer options for improvement. It’s called process mining.
There are a lot of ways to describe process mining: immediate, comprehensive, visual, objective. The focus is usually on how it extracts data from a system’s event logs, providing full transparency into how a business really operates. With insights gained from AI and machine learning, companies can then identify opportunities for process optimization.
Another way to look at process mining technology is through an analogy. In medicine, magnetic resonance imaging (MRI) technology gathers information from your body’s cells to create an image. Doctors then use that MRI visual to diagnose the state of your health. Process mining is similar, in that it gathers data from the most minute part of process activities and brings the pieces together to create a picture you can use to diagnose the health of your processes.
“Process mining bridges the gap between traditional model-based process analysis and data-centric analysis techniques such as machine learning and data mining.”
Wil van der Aalst, professor, RWTH Aachen University and Technical University of Eindhoven
Even the best documented procedures may differ from how they appear on paper to how they’re carried out in reality. As noted earlier, traditional techniques are not based on objective system data, and data techniques, such as data mining, do not address processes. Dr. Wil van der Aalst, deemed by many to be the “godfather of process mining” began exploring the possibilities of automated process discovery based on event logs in the 1990s.
This approach provides validity to the concept of process mining technology, as it merges the strengths of process and data-focused analyses. This is critical for businesses who want a true vision of where things are going right or wrong, because it uncovers issues businesses may not have thought they had or would have even thought to look for.
Returning to our earlier analogy, process mining is the evolution of process improvement methodology, just as MRIs are an evolution from x-rays in diagnostic technology. X-rays show disruptions in localized areas–for example, a broken bone–but don’t give a good picture of the surrounding supportive tissue, where other impairments or root problems may lie.
Process mining technology isn’t only about seeing where things can be improved, however. In addition to discovery, which is where you produce a process model from event log data without additional, non-empirical input, there are two other types of process mining: conformance and enhancement.
In conformance, an existing process model is compared with an event log of the same process to check alignment. In other words, confirming whether the process that’s being performed in reality conforms to the model process.
Enhancement extends or improves a model using data derived from an event log. Rather than uncover or compare process operations, the goal of enhancement is in its name–modifying or extending the existing model.
We tend to think of a model as the ideal scenario of how something functions. Yet in the context of process mining the end model should be viewed less as a fixed state and more like a map. The goal is to guide the user to the destination using the best possible route, knowing that things will change over time.
So, process mining technology uses a model with empirical data derived from the event logs in a company’s systems. But what is an event log, and how exactly does process mining change bits of everyday data into a fully-fledged process picture?
Most IT systems record every activity (“event”) happening in a process. Think of events as “digital footprints” left behind as people move through a process. Process mining technology captures the “footprints” from any number of systems throughout an organization and organizes them in a way that shows each step of the journey to complete that process, along with any deviations from the expected path.
This sounds straightforward, but there are many layers to consider. An event log consists of multiple cases. Each case has a unique ID assigned to it and consists of a set of events. These events are equivalent to the activities undertaken in a process. And each of these activities has a set of attributes.
Let’s translate this idea into an everyday example. Imagine you’re going to the supermarket. You shop, use your reward card and bank card at checkout, then leave. Straightforward, right? Yet during your trip, data is logged on your purchases at the fresh food counter (type of meat, bread, or cheese, amount or weight, time scanned), each of your purchases (time scanned, cashier ID, product ID), your bonus card (time used, number of points earned, your ID information) and your bank card (time used, location, total purchase, your ID information).
It is important to note that databases are not the only sources for this kind of data. Spreadsheets, business suites and other electronic logs can all be used to create a picture of a process. It’s no wonder, then, that before the introduction of process mining, it had been such a daunting task to make effective use of the world of big data to implement real process improvement.
Process mining technology doesn’t work by simply pulling data and organizing it. The rich visualization of process flow comes through the use of algorithms. An algorithm can even be viewed as a special type of process–a set of rules to be followed in calculations or other problem-solving operations. There are a number of existing algorithms that can be applied to the data to provide different insights into what’s happening.
Even the algorithms undergo scrutiny. To determine their applicability for process mining use, data scientists examine four factors (simplicity, fitness, generalization and precision). Rather than introduce additional variables, this feature reflects the flexibility of process mining. It means you can restrict or derestrict aspects of the results to look for specific attributes or show behaviors you need to modify to achieve your best possible efficiency.
From those first explorations of the potential of event log data in the mid-1990s, process mining technology has continued to evolve in its scope and sophistication. The expansion from academia into the commercial world has led to an extension of the applications of process mining into a wide range of real-world concerns.
One of the distinct advantages of process mining technology is the ability to work with data systems across an organization, with the potential to process endless volumes of data. Just as broadly, the underlying value of process mining is its ability to provide insights that enable process improvement, thereby increasing productivity and reducing costs.
That transparency has another benefit, in that it helps reduce risk in many concrete arenas. That enables CEOs to devote time to formulating scenarios and strategies for the “existential” concerns they also bear. More efficient processes and a game plan to advance a company’s agility is an attractive combination for any head of business.
Although conserving resources is almost a given goal of business, CFOs and other financial officers are also responsible for finding ways to increase revenue. By highlighting delays in payment processing that may cost the company discounts, or enhancing supply chain management to improve time to market, process mining can boost a company’s fiscal bottom line:
A chemical company was able to make a 20% improvement in both their on-time payments and invoice processing.
A telecommunications giant increased their perfect purchase order rate to 92% and improved their time to market by 20%.
An energy company saw $1 million in annual savings through supply optimization.
A company’s costs aren’t only related to lost revenue, loss of potential savings, or overhead related to salaries and supplies, however. Better efficiency also contributes to employee well-being and outlook, and the ability to leverage more substantive contributions from workers who aren’t tied up in rework or tracking down process hang-ups.
In that regard, an important component of keeping things on course in any business is its data system. Using process mining to find redundancies or instances of user error can have a significant effect on how well a system supports its business:
A high-end automobile components manufacturer saw a 30% improvement in their master data management process by identifying and removing a troublesome process step.
An industrial technology and infrastructure company increased efficiency by automating reports for production and assembly.
A grocery chain’s IT department was able to identify system issues immediately rather than waiting weeks to track and resolve issues.
Process optimization also facilitates agreement and adherence to the process. When things run quickly and smoothly there’s less need for people to try and circumvent the process to get what they want (e.g., the issue of maverick buying) and the checks and balances that are part of an effective process will also work to ensure process compliance.
Process mining technology can enable benchmarking across divisions or establish a benchmark for a certain process, department or division in a company. Again, the empirical nature of process mining ensures that benchmarks aren’t skewed by subjective reasoning or bias, and thus make any related goals or adjustments more easily accepted by stakeholders.
Another process mining benefit relates to the human factor. Specifically, how it can bolster internal cooperation. Operating from a common and objective understanding eliminates the need for discussions based on “hunches” or “gut feelings” that may lead to finger-pointing, but no effective forward movement. Data-derived results provide clarity so that planning can commence based solely on where opportunities actually exist.
And with process mining the opportunities do appear to be endless. For example, industry and infrastructure global giant Siemens has seen double-digit savings in the millions by applying process mining to their procurement-to-pay, warehouse and order-to-cash processes.
Finally, at the core of any optimization strategy is the goal of improving the customer experience. Happy customers are loyal, revenue-generating customers. Particularly in today’s world where social media feedback is playing a larger role in brand reputations and actions, process mining uncovers the bottlenecks that slow systems so you can avoid the kind of digital conversations that have costly repercussions:
A bank increased their retail throughput improvement by 30%.
A utility company reduced their customer onboarding time by five days.
A hospital reduced its emergency waiting room times by 56% in only 7 months.
Dr. van der Aalst and other process mining technology pioneers have said their vision for process mining is that it’s more than a technology tool; it’s a facilitator of discussions around processes and collaborations. In that sense, the true value of process mining is in changing not just the way your processes run, but how your business as a whole operates.
Let’s revisit the electronic distributor who was able to move from 24-hour to same-day service. Instead of wasting resources examining warehouse operations where they assumed there was an issue, to their surprise, their process mining analysis showed a breakdown in their sales process.
Once the source was identified it was easy to drill down to specific causes and address orders processed outside the system, and other self-created hurdles like supply locks or credit management. The company used process mining to set up a report on top of their existing HANA system to give them real-time updates, ensuring continuous implementation of their improvements.
Digital transformation initiatives can involve many types of systems and tools, and not all of them may interact. The key to success lies in finding a way to take a comprehensive look at what your business needs, evaluating what you can continue to use, what you need to upgrade or migrate, and what needs to be automated.
In spring 2018 Gartner released its “Market Guide for Process Mining.” In the guide Gartner noted, “some process vendors have shown that process mining can play an essential and fundamental part in digital transformation.” The report also recommended businesses invest in process mining before starting any automation initiative, whether at a task, workplace or process level.
Using process mining to get an overview of what’s going on within your processes, determining your path moving forward and monitoring after the fact can aid in everything from IT service management, shared services, and system migration to robotic process automation.
It can also address the issues identified by CIOs in the Deloitte survey. Using process mining technology at the beginning of the initiative establishes an objective review of what needs to be done to promote stakeholder buy-in, develop a plan, and facilitate confidence in both the plan and the training necessary to carry it out.
A common understanding further enforces compliance with the established and agreed-upon strategic vision and goals, and the speed and relatively low cost of process mining compared to consultants results in a significant ROI as the effort gets underway sooner and with more likely success.
The same understanding that aids in internal cooperation, efficiency, compliance and speed has an external effect. These are all things today’s customers have come to expect as a minimal level of service in modern business interactions, and the companies that meet or exceed expectations will be the ones that emerge as market leaders.
In light of predictions for the next couple of years–IDC’s assertions about AI support of digital transformation initiatives and the growth of digital strategies, and Gartner’s own prediction that the process mining market will triple or quadruple in that same period– businesses should position themselves now to make full use of the advantages process mining has to offer to keep them on the cutting edge.
There are many current options for process mining technology. Taking into consideration the estimates of IDC and Gartner, and the benefits of using process mining in digital transformation efforts, the marketplace is poised for rapid expansion. But if this is your introduction to process mining, how do you choose which vendor approach is the right one for your business needs and avoid hindering your optimization efforts?
The Gartner report not only provided background information on process mining but surveyed the current market of process mining providers. One of the key assessments was:
“With implementation in over 30 countries, over 80 consulting and implementation partners, and over 350 enterprise platform implementations (largest over 6,000 users, including large implementations in the U.S.), Celonis, a 350+ people organization, is clearly the market leader in process mining.”
The successes of the companies referenced in the previous section were all achieved using Celonis. They include mid-sized regional companies as well as international giants like Siemens and Vodafone.
The Gartner report further notes Celonis “focuses significantly more than its competitors on new innovative features.” These unique offerings include:
More than 300 apps developed through findings from over 1,000 customer projects;
An academy that provides comprehensive training for people in a variety of roles at various skill levels;
An extensive partner ecosystem for customer support; and
Advanced intelligent algorithms to identify areas of improvement and provide guidance toward root causes of deviations.
Celonis adheres to the expanded vision of the field’s founders, which is to think not just about software as a tool, but how the user can leverage that technology to transform the way their business operates. In other words, it’s not just about changing the technology people use, it’s about changing how people think about the positive impact of technology on how they operate.
We often acknowledge that change is hard, and organizational change may be harder still. But it doesn’t have to be. The transparency afforded by process mining can alleviate many of the fears and uncertainties about making process changes, whether they’re small improvements or involve moving in an entirely new direction.
Process mining technology offers the solution modern businesses need, from start to finish. It’s a powerful technology in an intuitive package that can support process enhancements from the start and continue to monitor performance for ongoing superior performance.
As it matures, the process mining field continues to innovate and exhibit endless use cases. As businesses of all sizes and across all industries search for ways to gain or maintain their competitive edge in an increasingly crowded and complex marketplace, the possibilities available using process mining will be essential to their long-term success and sustainability.
Hopefully you’ve gained a stronger understanding of process mining and its potential to transform the way businesses work. But you may still be wondering if it’s the right approach for your specific business challenges.
One of the key benefits of process mining is its flexibility. You determine the process and parameters for analysis, then plan for the changes you’re able to implement. Whether your goal is increasing efficiency or implementing automation, major change doesn’t have to be overwhelming, and you don’t have to go it alone.
Celonis offers several options to assist you with your next steps:
Do a 30-day trial of Celonis Process Mining. It’s browser-based for quick and easy install, and includes tutorial, tips and a personal demo space to see your own data in action.
Join one of our daily 30-minute webinars or schedule a personal instructional session. Ask questions of our Data Scientists to learn more about what you’re seeing and how to best use process mining to your advantage.
Contact a representative for any other questions, or to get started right away on the process mining solution that’s right for you.
Celonis is the leader in business transformation software, turning process insights into action with process mining technology. Its Intelligent Business Cloud allows global organizations to guide action and drive change in the operational backbone of their business, resulting in millions of dollars saved and an improved experience for their customers.