Gartner. Licensed for Distribution Market Guide for Process Mining Published 11 November 2021 - ID G00737056 - 58 min read By Marc Kerremans, Tushar Srivastava, and 1 more Recent trends in autornation and knowledge of the underlying processes and interactions are key to digital transformation. For data and analyt and stronger performance on an organizations most critical priorities, leaders, process mining delivers insights that enable faster, smarter decisions Overview Key Findings «Traditional interview-based process discovery and modeling is costly and time-consuming because of gaps in business knowledge, a lack of ‘objective validation techniques and poor formalisms, 1 Formal standard operating procedures, policies, work instructions or best practices that are baked into enterprise applications are often ‘compromised by informal behavior that bypasses governance. += The business value of process automation initiatives is often unclear, The assumption is that the actual operational data is available only after the automation, leading to a business case based on estimated or poor operational data and an initiative with limited sustainable business value, '= Creating adaptability and resiliency in an organization is about having the capabilities to adapt to multiple experienced or expected “stressors” to remain a healthy organization. But as with all capabilites, it needs operationalization within the everyday organizational realty — the how. Recommendations ‘To drive business transformation and optimization through technology innovation, data and analytics leaders must: '= ely on proven methods, techniques and tools that, based on available day-to-day operational data, provide an accurate model of the ways of workin a format that can be understood by anybody in the organization, This will improve visibility and understanding ofthe actual performance of business operations and processes. Engage all process participants and create awareness with business and operational colleagues on the benefits, capabilites and use of process mining by introducing small, short-term pilots. This will create valuable insights and overcome future shadow processes, ‘= Make explicit the extensive and expensive set of business processes underpinned by a patchwork of technologies that are often not ‘optimized, lean, connected or consistent. This wil deliver a path to digital operational excellence, addressing the tremendous collective debt within organizations «= Explore use cases that go beyond traditional, internally focused process discovery, process validation and model enhancement by targeting business operations and interactions with external parties, such as customers, This “customer journey mining” attracts an entire new ‘audience, and it aligns customer journey maps with internal process maps. Market Definition Process mining is designed to discover, monitor and improve eal processes (not assumed processes) by extracting knowledge from event logs readily availabe in today's information systems. Process mining includes automated process discovery (i. extracting process models from an event log), conformance checking (i.e, monitoring deviations by comparing model and log), social network/organizational mining, automated construction of simulation models, model extension, model repair, case prediction, and history-based recommendations. '? Market Description ‘The starting point for any process mining task is an event log. Each event in such a log refers to an activity (Le.,a welhdefined step in some process) and is related to a particular case (ie, process instance). The events belonging to a case are ordered and can be seen as one “run” ofthe process, The sequence of activities executed for a case is called a trace. Figure 1 reflects a simplified representation of the basle process mining technique. ‘These events need not be present in a readily available lag file from an enterprise application, The advantage of having this application logfile ‘available is that the events are already structured, Events can come from anywhere (e.g., desktop applications, loT devices and documents), and in that case, these events must be structured to be ready for ingestion by the process mining engine. Figure 1: Basic Process Mining Technique Basic Process Mining Technique Real-World Activities Cu arc Cd Notify o . Go o 5h OO 0 —ee or] 367 Analyze Purchase Create Direct Request eo welt {482 {267 case 9700 20.0819 11:45 Completed ee Send RFQ te Suppior caseso | a10ese1200 | Ontos ‘81 Gartner Based on welhsupported constructs in academic research, these techniques have moved beyond the boundaries of academia. Software vendors have packaged the techniques in easy-to.ise graphical toolsets. Sometimes, the vendors provide these toolsets with additional services {featuring opportunity detection, implementation interpretation and best-practice approaches, In 2008, the Institute of Electrical and Electronics Engineers (IEEE) established the Task Force on Process Mining, and in 2011, it published a ‘manifesto to promote the topic of process mining. In the context ofthis task force, a group of more than 75 people involving more than 50 organizations created the Process Mining Manifesto. By defining a set of guiding principles and listing important challenges, this manifesto serves as a guide for software developers, scientists, consultants, Business managers and end users. The goal sto increase the maturity of process mining as a new too! to improve the redesign, control and support of operational business processes. [At approximately the same time, one of the fathers of process mining, Professor Wil van der Aalst, published the first book on the subject. In 2016, an updated and extended version of the book appeared, titled "Process Mining: Data Science in Action.’ Van der Aalst also presented successful massive open online course (MOC), with more than 143,000 people from more than 190 counties participatingWe have included responses from our interview with van der Aalst (see Note 3) discussing the process mining market. The biggest value ‘accelerator for process mining was the collaboration of process mining vendors with wellknown enterprise applications, such as Oracle, Salesforce and SAP. These vendors have heavily promoted process mining to improve processes underpinned by the applications and of the applications themselves. Finally, process mining has moved into areas other than process discovery, such as conformance checking, productivity improvement, customer imeractions and social networks. It has even spread into areas ofthe Internet of Things (loT), manufacturing and logistics distribution networks, Which have demonstrated the sustainable-value-creating capablities of process mining. Recently, we have seen evidence of process mining Providing one of the cornerstones of a companywide digital transformation initiative Gartner started to report on process mining in 2008 as automated business process discovery (ABPD). We pointed out that ABPD can ‘complement traditional business process discovery, help a process analyst better understand the dynamics of business processes, and improve the quality of process models, consisted of two steps: ‘= One phase focused on process selection and prioritization, which clearly sets the improvement goals ‘= A-second phase focused on process information capturing [At that time, we argued that business modeling has to precede business process modeling. Determining why an organization performs certain activites is often neglected between scoping process improvement initiatives and beginning process modeling and analysis. Developing a ‘complete business operations model based on the optimization of value and asset utilzation can provide a reference model for business processes, Such @ model indicates where enterprise value is crested in different parts of the organization and how high-level processes affect that value creation. This wil lead to quicker savings and more long-term business value than the traditional approach of first creating a detailed business process model During the past few years, we have had extensive interaction with end-user clients, have had many contacts with the academic researchers (yearly International Conference on Process Mining) and have conducted a survey with most ofthe process mining vendors. Given this background, Gartner has identified 10 capabilities for process mining for 2021 1 Models of processes, exceptions and process instances (mostly referred to as “cases), and employee interactions — Automated discovery of process models, exceptions and process instances, together with basic frequencies and statistics '= Support for customer interactions, customer journey maps and related analysis — Automated discovery and analysis of customer interactions, 2s well as alignment with internal processes ' Conformance-checking and gap analysis capabilities — Capabilties to check conformance not only graphically through ovetlays, but also through data analysis and performing gap analysis f= Intelligent support for process model enhancement ~ Enhancing or extending existing or a prior process models by using additional data from the recorded logs and events ' Data preparation and data cleansing support, supporting big data — Different ways to handle data ‘= Real-time dashboards with support for key performance indicators (KPIs) that are continuously monitored and enable decision support — Realtime or near-real-time connections to continuously monitored and adapted KPls in dashboards for specific oles n the organization ' Predictive analysis, prescriptive analysis, scenario testing and simulation — Advanced analysis capabilities that use contextual data 1 A platform that extends the process mining capabilities across different processes with advanced analytic capabilities and decision ‘management capabilities and that also offers APIs to create process mining apps ~ Allowing organizations and partners ofthe process ‘mining vendors to create applications, such as financial auditing tools += Task mining — Inferring useful information from low-level event data available in Ul ogs. These Ul logs describe the single steps within a task done by a user — for example, when using a workstation ~ based on keystrokes, mouse clicks and data entries, 1 Execution capabilities that turn “insights” into “action” ~ These capabilities could range from simple updating source applications {applications that delivered the events for process mining) to creating scripts that support execution of tasks. Data and analytics leaders should realize that not all process mining vendors will support all features. Because of the emerging nature of some of the use cases and the underlying discipline, many suppliers have gaps in their process mining offerings. Some of these process-elated capabilities and features may be available as part of other types of products, such as enterprise business process analysis platforms and‘automated testing tools, However, here we consider only vendors and tools that provide process mining as a stand-alone offering that Is capable of adcressing all or most ofthe use cases. Market Direction Adoption Gartner sees five main rivers for the adoption of process mining ' Digital transformation Axtificial intelligence = Task automation Hyperautomation = Operational resilience Digital Transformation Digital transformation drives growth in business users’ awareness ofthe benefits of analyzing and understanding their own processes and business operations in a broader enterprise context (see The Business of Digital Business). This is occurring because digital business and digital transformation have become major themes, and because processes are important constituents inthe operationalization of these digital business initiatives. Furthermore in this era of digital business, data and analyties leaders need a way to reflect on how these new technological capabilities can provide value tothe business and, ultimately, to the customer. Again, its here that analyzing business operations, processes and customer interactions can show how and where to activate these capabilities to create business value, The initial, tical discipline of BPM provides, from a business perspective, the visbilly needed to gain a shared business understanding of the process. This understanding can become a springboard to making processes more effective and allowing them to be changed more rapidly. In enterprisewide digital transformation initiatives, its important to align and adapt these processes with client interactions to attain the targeted business outcomes. Artificial Intelligence Algorithms and several waves of artificial intelligence (Al) have been at the core of computing for decades, However, the ability to develop algorithms that act on vast amounts of data to identify patterns creates new opportunities (see 2022 Planning Guide for Analytics and Artificial Intelligence). Inthe digitalized economy, business algorithms deliver new sources of value in business ecosystems, With the use of artificial intelligence and advanced machine learning algorithms, data acquires meaning, and new and powerful insights can be derived from it ‘A powerful example of this data science in action, process mining shows how algorithms can be used as a mechanism to capture knowledge ‘and insight in @ packaged form that can be simply reused in a consistent fashion, Data and analytics leaders must focus more precisely on those instances in whch algorithms are providing the insight and have become pivotal o competitive differentiation. Task Automation (RPA) ‘The current hype around task-level automation, such as robotic process automation (RPA), has resulted in large one-off cost savings (see Magic Quadrant for Robotic Process Automation). Usually ths is related to specific conditions in which people work on repetitive, data-centric tasks, ‘mostly because ofthe lack of APIs fr old legacy systems. Preferably, these tasks are embedded in processes that are fixed for the next five to 10 years, However, in most cases, tasks are part of processes and operations for which change is the most common characteristic. By accurately assessing the processes to which these tasks belong, we can identify “hot areas" in the organization where alot of effort is wasted in repetitive tasks. Then, we can see whether these tasks can be partly or fully automated via RPA. This is where process mining can complement RPA perfectly to offer a wider context and help implement this task automation. This results in long-term sustainable business value and averts the shortcomings of a short-term perspective focused on large, one-off cost savings. Hyperautomation ‘The current business reality is that organizations have a tremendous amount of “collective” debt (technical, process, data, architecture talent, security and social) that significantly impacts their value proposition, The root cause is an extensive and expensive set of business processes underpinned by a patchwork of technologies that are often not optimized, lean, connected, consistent nor explicit. ‘The collective impact of these business andIT realties is an extensive number of initiatives (often disparate and siloed) almed at applying some type of automation across the fabric (Le, the DNA) of knowledge work fo ether efficlency, efficacy and/or business agility. The dilemma is that the term “automation” has been used for over five decades. Therefore, Gartner is using the term “hyperautomation’ as a way of creating @ “common currency" for discussion to showcase that there is a need to orchestrate al islands of automation, new and existing, within an‘organization (see Figure 2). Additionally, i's important to emphasize that hyperautomation is not about technologies (products or services) Rather it is @ design pattern (see Move Beyond RPA to Deliver Hyperautomation), Process mining is a fundamental pat in creating visbilty and understanding before you automate, It also visualizes how diferent islands of ‘automation are connected and how continuously implemented and connected automation can be improved through its monitoring capabilties. Figure 2: Hyperautomation z Hyperautomation Wetec oe 3 z Coa Tenet ete) i perati Lica 3 process Creo 3 en 4 as Oe Narrow Broad Scope of Automation Gartner Operational Resilience Operational resilience is a set of techniques that allow people, processes and information systems to adapt to changing patterns. tis the abilty to alter operations in the face of changing business conditions, Operationally resilient enterprises have the organizational competencies to ramp Lp oF slow down operations in a way that provides @ competitive edge and enables quick and local process modification (see Figure 3). Figure 3: Operational Resilience £ nce Operational resiliency enables enterprises to Seek ry Model > Adapt adapt by: + Establishing relevant and achievable goals Pee) Tce od Ee een he PC oa ers + Breaking down complexity Cea cary ia + Facilitating collaboration + Creating organizational capability for resiliency + Defining measurements and adjusting Gartner.Operational resilience uses information from “seek” and “model” to enable the third ofthese disciplines, “adapt” To enable enterprises to adapt, operational resilience uses output from the "seek" and “model” disciplines and returns information that serves as feedback that improves these ‘activities as well (see How Process Mining Can Support Operational Resilence in Times of a Crisis), The techniques underlying process mining provide a new and enhanced way to encompass the sense and model capabilites Based on available day-to-day operational data, process mining continuously seeks and finds the relevant objective operational data, The ‘advanced process mining algorithms then provide an accurate model ofthe ways of work ina format that can be understood by anyone in the organization. This ensures that everybody can be engaged in the change initiative. Furthermore, i allows for continuous adaptation and Improvement because, after the adaptation, the new operational data will give insights into the new adapted way of working ‘Autonomic Operations and Autonomic Business ‘An Interesting evolution inthis thinking Is revealing Itself when we start addng this adaption ~ that is, execution through automation ~ tothe capabilities of process mining. This would allow the fullzesilence life cycle to be handled nan autonomic way: autonomic operational resilence. Of course, ane necessary component would be decision management support. Utimstely this autonomic operations and operational resilience could lead the way 0 autonomic business by extending the operations tothe organizational context within a business operations model Five Common Use Cases for Process Mining Process mining can be applied indifferent areas of an organization. Therefore, it can have many use cases, and a multitude of stakeholders may depend on the use case and the scope of the process mining initiative. The process mining market can be segmented along fve types of use ‘cases, each of which has its own stakeholders, roles, objectives, and related features and functionality. In short these use cases cover: 1 Discovery and analysis ‘= Auditing and compliance = Automation Digital transformation IT processes. We algo indicate possible future directions for each use case. At the end ofthis Market Guide, we interview Professor Wil van der Aalst to get his perspective (see Note 3) Improving Processes by Algorithmic Process Discovery and Analysis Process mining provides visibly for and understanding of actual business operations and processes by applying a set of algorithms to events, resulting in highly adaptable, highly maintainable an¢ highly validated process models. Sesides identifying process inefficiencies, this technique delivers insight into where to improve operations (for example, ina digitalization initiative) and how to attain targeted business outcomes, By supporting process efficiency and effectiveness, process mining tools are key enablers of process improvement intiatves and their rek disciplines. n this use case, data scientists typially support process improvement teams to discover, analyze, improve and optimize processes Using methodologies such as Kaizen (lean); Plan-Do-Check-Act (total quality management); or define, measure, analyze, improve and control (OMaIc [Six Sigma). Process mining is highly complementary tothe traditional business process analysis area, enabling business process analysts to document, analyze, streamline, and redesign complex processes at a conceptual and logical level, regardless of any planned automation (see Market Guide ‘or Enterprise Business Process Analysis) Direction ‘Most of the new capabilites in support of this Use case are found inthe data-entry functionality (data collection, cleansing and preparation), the segmentation of cases (process instances) and predominantly on the analysis side. Due to the recent advances in Al and machine learning, the process analysis capabilties are extended into areas such as predictive analysis, prescriptive analysis, scenario testing and simulation Another evolution is the support for the interaction of different processes (not instances of the same process) and how process instances (ot ‘cases) interact with each other, This evolution is enabled by looking at events supporting multiple processes, and by creating event-relationshipstructures or diagrams. Together with the execution capabilities that turn insights into action, this capability has been labeled as most differentiating by the participating vendors in our survey. Furthermore, if process mining tracks clients and ther interactions — ae wel as their touchpoints withthe organization as the main object rather than an order invoice or request ~ then this can be seen as “customer journey mining” It pens up an entire new audience. Finally, we see a promising direction in applying the process mining techniques to objects other than work objects going through different tasks, such as products, services, resources, channels and clients, Combining all these business model constituents, we see the emergence of a new area that could be labeled “business mining" or ‘organizational mining,” One of the frst use cases isthe exploration for how customer Interactions align with internal business operations and processes. This analysis helps organizations address how they can actively impact customer experience and customer retention through intemal operational improvements. Improving Auditing and Compliance by Algorithmic Process Comparison, Analysis and Validation “Most organizations have internally defined standard operating procedures, policies, work instructions and best practices baked into enterprise apalications, such as ERP and CRM. In practice, many ofthese predefined operations are often complemented by "shadow operations. Process ‘mining helps validate or audit whether actual operations are in conformance with these defined operations, More accurate assessments of process deviations and compliance issues, such as segregation of duties, help manage these risks and communicate findings in an audit report. This also helps ensure the reliability of external financial reporting and compliance with external laws and regulations. in these cases, the stakeholders of process mining can be found in financial management and in risk and compliance teams. ‘Another variation of this use case isthe comparison and analysis of the same processes across different business units, subsidiaries or even countries (i.e, comparative process mining) Direction Large, external auditing companies have applied process mining techniques in their auditing services. Most of them have been experimenting {for a while, However, some of them have recently partnered with process mining vendors to build platforms on which they can run their auditing ‘activities and reports in a more automated way. Lately, we have noticed a trend toward the use of the mined information in a more active way. By comparing what is actually going on in real time, such as the allocation ofa particular task to a process participant and the insights delivered by process mining, these participants can be ‘uided in completing the task. Or they can be urged to prioritize certain tasks in order to avoid further expediting of the process instance. This ccan be done by simple alerting or more advanced workflow capabilities, also guiding the next steps in the process. Yet another evolution is going beyond alerting to actually changing or executing next steps in the process, An example is the change of delivery plans based on the actual situation, combined with the patterr-based insights, in order to deliver on time Improving Process Automation by Discovering and Validating Automation Opportunities Process automation initiatives have been delivering highly improved efficiencies, but they have resulted in limited sustainable business value because ofthe assumption that the actual data is available only ater automation has been completed. This has led to less accurate process designs as input forthe process automation. Process mining delivers the actual operational data before running into automation and supports ‘making a more accurate business case on the Improvement of business outcomes. Process mining will help business analysts discover and ‘assess the opportunities for process automation. Quantifying the value of automating a process (or task) ig huge. It effectively creates the business case for automation, whichis often difficult for some clients to create on their own. Tasi-level process automation, such as RPA, has often missed the process context and has not provided awareness of the changes that occur to the processes in which these tasks are embedded. Process mining will help in two ways: frst by discovering the opportunities for task ‘automation in a mare holistic process context, and second by connecting automated tasks to its process context so that business process changes can be anticipated and reflected in these tasks. Finally, process models are often used to configure enterprise applications and their integrated processes, such as ERP and CRM, However, traditional process models based on interviews often provide only a limited or biased picture of the truth, and highly abstract resus can be interpreted in many ways. Process mining ~ as a complementary approach to traditional business process modeling — provides process Visibility by generating highly adaptable, highly maintainable and highly validated business process models, ction Because of the huge hype around RPA, which is task automation, process mining draws alot of attention from the RPA community. Process ‘mining improves the success rate of task-level automation, such as RPA, through visualizing and understanding the process context so that, when processes change, the automated tasks can adapt without losing thetr relevance. Moreover, process mining helps spat and prioritize opportunities for task-level automation, We sil see several partnerships being established by the most wellknown RPA vendors and some of the process mining vendors.‘Some RPA vendors, as well as a couple of new companies, have developed capabilities that apply neural networks and deep learning to data captured from screens and desktops to derive the process of how an end user completes a task, Ahough these vendors call this ‘process discovery” this should not be regarded as process mining, but rather, a use case of “task mining” (see Innovation Insight for Process Mining and Task Mining). It captures the process of a task based on keystrokes, mouse clicks, opening and closing of applications, and entering data into fields. This is valuable capability that can even automate the configuration of RPA bots, whichis highly complementary to process mining. ‘Another promising trend isthe inclusion of operational technology (OT) int the collection of event logs by using Internet Protocols (or loT). The Combination of loT events and other events will support the discovery and validation of the value of enabling things to cooperate in processes. Earlier, we discussed hyperautomation and the relevance of process mining within the toolbox of technologies supporting this hyperautomation. Finally, we have seen an interesting direction of building complete application functionality based on process mining results. By discovering the ways of work and using this mined information to automate the sequence and guidance of these different tasks, process applications are being scripted and built. Combining this capability with continuous monitoring and rediscovery ultimately leads to the creation of continuously adapting or dynamic applications. This Is also an element of autonomic operations (see the five main drivers of process mining in the Adoption section), ‘Supporting Digital Transformation by Linking Strategy to Operations, and Creating a Resilient Organization I we shift the scope of process mining to the operational and organizational level, itis a small step to link these operational insights to big strategic initiatives like digital transformation or digital business, This is a growing use case that is getting mote attention because of the themes of digital business and, more recently, digital transformation, This use case provides visibility, analysis and understanding around business operations models *that represent a way of doing business by providing (near) realtime information to all end users on how they are performing and what could be Improved. In some cases, it has been reported that it even helps end users think about digtalization opportunities. ‘This use case can be regarded as the combination of three markets (enterprise business process analysis [EBPAl, digital twin ofthe organization [070] and process mining) that Gartner has been deserbing in separate Market Guides Direction The use case around linking strategy to operations or focusing on business change and digital transformation actually delivers @ component of the concept we have described as DTO. In that way, process mining grows into an essential part of any business transformation initiative. It helps guide digital business initiatives withthe data connection part ofthis DTO (see Optimize Digital Business Transformation by Creating a Digital Twin of Your Organization and Market Guide for Technologies Supporting a Digital Twin of an Organization). This use ease has shown that process mining from an operational perspective can radiate into the wider end-user community in an organization ‘The combination of these technologies has been touched upon in recent research around DigitalOps (see DigitalOps Helps Connect Business “Models to the Digital Business Platform), Leading organizations are carefully managing the different process models and past performance data. Combining these process and decision models (see Develop Good Decision Models to Succeed at Decision Management) withthe situational context of work helps control and optimize the execution of work in realtime. It's this combination — the blending of process ‘modeling and performance monitoring to support direct execution and real-time optimization of the digital business platform — that Gartner calls “DigitlOps.* Process mining can provide these process models and measurements continuously and accurately (see Figure 4). Figure 4: Digitalops Combines Three Domains leigitalOps Combines Three Domains Process Modeling Process Monitoring Planning and Knowing What Happened Understanding Work + Business Operations + Process and Decision Monitoring + Data and Events + Business Performance + Goals and Motivations Dashboards + Machine Learning + Analytics, BAM and BI + KPls and KRIs Modeling Monitoring Process/Task Execution Supporting and Driving Work + Orchestration and Coordination + Process and Decision Execution + Event Management + Algorithms and Bots + Integration Interfaces pro Note: Organizations are moving to combine these domains, rather than dealing with them separately, tal twin ofan organization Gartner Finally, we see process mining as instrumental toa bigger trend thatis currently in its infancy — namely, autonomic operations and autonomic ‘operational esifence. Operational resilience isa set of techniques that allow people, processes and informational systems to adapt to changing patterns. It isthe ability to alter operations inthe face of changing business conditions. Operationally resilient enterprises have the ‘organizational competencies to ramp up or slow down operations in a way that provides a competitive edge and enables quick and local process modification (see Figure 3) Process mining and the evolution of Al and ML count forthe autonomic component. The techniques underlying process mining, and eventually business mining, provide @ new and enhanced way to encompass the sense and model capabilities of operational resiliency, whereas the Al, ML and scripting capabilites provide the adaption or execution capabilities of operational resilience. Improving IT Operations Resource Optimization by Algorithmic IT Process Discovery and Analysis This use case is not a direct goal or result of business-outcome-drven process improvements. However, the discovery, analysis and optimization of T operations and processes is a new use case that can certainly deliver @ lt of benefit to IT organizations and widely adopted practices such ‘as DevOps. Consider its use in the improvernent of development and testing processes, preparation of migration processes, system error diagnostics, technology usage, and postimplementation/transformation stabilization. This stabilization centers on usabilly Issues, client service delays and reproduction scenarios, Some companies are using process mining to check the use of certain application functionalty, and to determine which functionality is a good candidate for modernization a rationalization Direction Although this use case emerges recently, we expect alot of uptake on this specific use case, especially rom vendors and system integrators that have strong T backgrounds. Many connections to IT service management (ITSM) offerings — such as Atlassian, BMC Software, Broadcom (CA Technologies) and ServiceNow ~ reinforce this category of use cases, Some of these vendors have started to introduce their own process ‘mining functionality within their current offerings, while others have deep partnerships. Market Size 1p 2020, Gartner estimates the process mining market for new product license ané maintenance revenue to be $550 million, which was over 70% ‘market size growth from the previous year (see Forecast Analysis: Process Mining, Worldwide). The process mining market is forecast to keep‘growing between 40% and 50% and will pass $1 billion in 2022 A large follow-on market also exists for consulting and services in implementing these tools and the methods for using them. When calculating process mining market revenue, we did not include consulting or service revenue. For some vendors, this addon consulting and service revenue significantly exceeds thelr software revenue, Market Analysis Ina short market survey, we questioned the process mining vendors about the adoption and importance ofthe diferent basic types of process ‘mining (see Process Mining Manifesto ") and the adoption ofthe different scope areas, We also asked for an assessment of ther future expectations on both topies. The survey results on basic process mining types confirmed that, today, most implementations stil focus on discovery. However, we see a significant trend toward an increased focus on enhancement process mining type, whereas conformance is stabilizing (see Figure 8). In this chart, we combined the data ofthe surveys from previous years’ Market Guides with this year’s survey data. For 2017, 2018, 2019 and 2020, these are observed values. The values for 2022 are expected values. Figure 5: Adoption of Basic Process Mining Types + Adoption of Basic Process Mining Types Adoption Basic Types (%) 207 2018 2019 2020 2022 estimate) 60 30 0 Discovery Conformance Enhancement Gartner [Another question in the survey asked for the predominant use cases in which process mining has been applied. The data confirms that most process mining initiatives in 2019 targeted business process improvements. However, in 2021, we have seen an increase in use cases targeting process automation — business process automation (BPA) and RPA — and use cases targeting digital transformation (see Figure 6) Figure 6: Process Mining Use Cases leProcess Mining Use Cases Use Cases (%) M2018 2019 M2020 MH 20Z1(estimate) | 2022 (estimate) 50 a 37 32 25 a 23 23 2 25 2 19 2 . 7 8 7 15 16 2” 56 ‘I 0 a Business Process Auditing and Process Digital IT Improvement Compliance Automation Transformation Operations Gartner Representative Vendors ‘The vendors listed inthis Market Guide do not imply an exhaustive list. This section is intended to provide more understanding of the market and its offerings. Market Introduction Vendor anavy Apa (Lana Labs) Apromore Businessoptix Table 1: Representative Vendors in Process Mining Product, Service or Solution Name J, Teneline LANA Process Wiring, LANA Cernect ‘Acrorore rterprise Eciton inessCotix CColons Execution Managemert SystemVendor Datreks EverFlow Flxeon Integris Livejourney Minit Process Analyties Factory (PAF) Puzzle bata ‘PR Sofware SAP (Signavio) Software AG Soroco Stereotocic uipath See Nete1 Source: Garner (November 2021) Vendor Profiles Product, Service or Solution Name Datricks EverFlow Disco IBN Process Mirirg lore Process Livejourey Mii PAFnow PreDiscovery (PR >rocessanalyaer ‘SA® Signavio Process Intelionnce ARIS Process Wining Scout Pltfor ‘SterecLOGIC Frocess Mirirg, StereoLOIC Task Minirg Path Proce Mining, Let Task Virirg The survey revealed the use cases that each vendor focuses significantly more on than the average of all vendors. Keep in mind that all vendors, ‘active in the process mining market provide features and functionality for most of the use cases. However, the focus, breadth and depth of these features and functionality will vary depending on the predominant use case marketed and implemented by the different vendors. ABBYY Based in Milpitas, California, ABBYY delivers Timeline version 5.1.0:1 The acquisition of Timeline by ABBYY and the addition of Task Mining to Timeline positions the company to address case-based digital transformation initiatives. ABBYY provides end-to-end visibility into highly variable content-centric processes and how they are performing to identity challenges, improve efficiencies and enhance customer experiences. ‘= Timeline is purpose-built to handle unstructured, case-based, ad hoc processes in addition to structured business processes. Timeline uses a patent-pending methodology that is fundamentally different from any other process mining application commercially available today. Coupled with ABBYY's Content Inteligence solutions, this is extended to include insights from unstructured and semistructured document types found in case-based business processes. = Timeline automatically monitors process performance in realtime to assist in triggering remediation processes or RPA technologies (OEM partnership with Blue Prism). After analyzing al of the data, Timeline technology observes event executions and predicts the next best action, = According to our survey, ABBYY focuses more on business process improvement and audit and compliance use cases. Our survey revealed that it focuses significantly more than its competitors on modeling of processes and exceptions capabilities, followed by conformance checking and gap analysis capabilties, based on its monitoring capabilites. Appian (Lana Labs) Based in McLean, Virginia, Applan provides LANA Process Mining and LANA Connect version 4.12: ‘= Appian acquired Lane Labs, integrating the LANA Process Mining toolset into the Appian LowsCode Automation Platform. Appian delivers bidirectional integration with LANA Process Mining, allowing discovered processes to be used to jump-start automation projects in Appian ‘and allowing Appian process analyties data to be fed into LANA Process Mining for greater insight into enterprise processes. = LANA Process Mining has @ high degree of analysis automation — with automated root cause analysis and improvement recommendations: based on machine learning — and automated actual target analysis against BPMN. LANA Process Mining has a strong focus on integration, ‘exemplified by its open REST APIs and multisource transformation, as well as by the introduction of a separate ETL tool for process mining (LANA Connect) ‘= Beside its own prediction algorithms, LANA Process Mining provides a customization layer that extends LANA with all kinds of ML/AI algorithms. These predictions will be refreshed on each new data upload and fiter setting provided, ‘= According to our survey, Lana Labs focuses more on audit and compliance and business process automation opportunities use cases. Our survey revealed that it focuses significantly more than its competitors on data preparation and data cleansing supporting big data, as well as process model enhancement capabilities. Apromore Based in Melbourne, Australia, Apromore provides Apromore Enterprise Edition version 7.21 ‘= Apromore is a commercially supported open-source process mining tool that implements the full spectrum of process mining functionality — ‘automated process discovery, model-based performance analysis, variant analysis and conformance checking, and realtime predictive process monitoring, Apromore has more than 30 consulting and implementation partners to deliver process mining solutions globally '= Based on an advanced algorithm for process discovery (Split Miner), Apromore can instantaneously discover Business Process Model and Notation (BPMN) models for interactive analysis. Apromore thus supports switching in one click between a “process map" view, which shows, ‘an informal view of dependencies among tasks in the process, and a BPMN view. 1 Apromore's predictive monitoring engine uses ML models to produce predictions in real time, based on event streams. The solution includes ‘an open API integration and a no-code extraction transformation and loading (ETL) environment to define and schedule data extraction and transformation pipelines via a graphical UL ' According to our survey, Apromore focuses more on business process improvement and on improving process automation use cases. Our survey revealed that it focuses significantly more than its competitors on conformance checking and gap analysis capabilites, followed by predictive analysis, prescriptive analysis, scenario testing and simulation capabilities BusinessOptix ‘Based in Kansas City, Missour, and London, BusinessOptix offers BusinessOptix version 7.0; 1 BusinessOptixis an ontine platform that models an organization's multiple layers from high fevel organization design and cross functional ‘mapping to process maps and dacumentation. BusinessOptix appeals to large consulting and system integrators, business process ‘outsourcing, and end-user organizations that are creating a center of excellence (COE) or intellectual property (IP) at scale on the platform.1 BusinessOptixis an integrated platform that includes process mining alongside other capabilities (including process modeling, process ‘analysis, scenario modeling and simulation, operating models, work instructions, and transformation dashboards and maps) to support ‘organizations understanding and improving their processes at departmental, functional and organizationwide levels. ' BusinessOptixis one of the first vendors offering process mining capabilities within a DTO platform. Furthermore, BusinessOptix optimizes automation efforts by taking mined data, creating proposed automation solutions to demonstrate business benefit through cost or throughput, and providing insight into future capacity planning needs. ‘= According to our survey, BusinessOptix focuses more on digital transformation and on business process improvement use cases. Our survey revealed thatt focuses significantly more than its competitors on predictive analysis, prescriptive analysis, scenario testing and simulation capabilites, followed by conformance checking and gap analysts capabilities. Celonis Based in Munich, Germany, and New York, Celonis offers the Celonis Execution Management System (EMS) + Celonis is a global company with 16 offices around the world and serving ever 1,000 clients. Ithas more than 200 consulting and implementation partners. Furthermore it has huge customer projects ~ the largest is more than 8,000 users, and the largest data model Contains over 500 million cases and more than 10 billion events, From a market share perspective, Gelonis is dominating += The Gelonis Execution Management System is @ containerized, microservices-based cloud platform. Beyond all the common process mining functionality, this cloud solution provides real-time data ingestion, including streaming data (from the Lenses.io acquisition), task mining of ser behavior, machine-learning-based process modeling and simulation, and process-inteligence-driven action flows that proactively steer ser actions or directly automate actions in source systems, "= Celonis currently offers execution apps for opportunity management, order management, inventory management, accounts receivable, procurement and accounts payable. Additionally, a growing number of partners are offering their own execution app within Celonis’ EMS Store ‘and thereby offering a way to provide thelr expertise to all Celonis customers ' According to our survey, Celonis focuses more on digital transformation and business process improvement use cases. Our survey revealed that it focuses significantly more than its competitors on new innovative capabilities, such as capabilities supporting the interaction of different processes, and execution capabilities, exemplified by its application portfolio Datricks Based in New York, Datricks offers Datricks version 2.1 ' Datricks, one of the new entrants in the process mining market, uses process mining and explainable Alto automatically map internal finance ‘and other compliance+elated processes and to identify risks in real time. Combining the world of governance, risk and compliance (GRC)/integrated risk management |RM) with process mining and A, in realtime, Datricks transforms the way internal auditors approach compliance programs, ' Datricks includes a comprehensive library of realtime contrls, allowing continuous audit and compliance and enabling Sarbanes-Oxley Act (SOX) readiness, Focusing on this intemal auditing aspect, and on detecting process anomalies using AI/ML techniques, Datricks can be seen ‘as a "business process firewall" that protects processes against internal fraud and compliance risks. 1 Datricks exposes a ‘process warehouse” to 8 tools, which means the BI tools access a universe describing a business process model rather than the underlying technical database tables. In that way, Datricks can be incorporated into a company's existing BI tools and platforms, enabling the company to integrate process mining insights within its strategic BI dashboards. ‘= According to our survey, Datricks focuses more on audit and compliance anc on business process improvement use cases. Our survey revealed that it focuses significantly more than its competitars on conformance checking and gap analysis capabilities, based on its internal ‘auditing functionality, followed by real-time dashboard capabilities, EvefFlow Based in San Francisco, California, EverFlow offers EverFlow version 3.1 1 EverFlow was born as a spinoff of Icaro Tech, a Brazilian intelligent operations product and service provider, and aims to democratize the use ‘of process mining. EverFlow mixes advanced process mining and machine learning techniques with big data to buld a platform that anyone ‘can use to analyze and improve business and technical processes, = EverFlow uses the combination of filtering, customized dashboards, ciferent visualization perspectives, drilldowns and partitioning (machine learning algorith), among others, to explore process-elated data. Furthermore, it uses advanced mining algorithms, such as an Inductive Miner drectly-fllows (IM) variation to generate process trees and BPMN models, even in big data scenarios,' nits effort to support big data, verFlow applies machine learning clustering algorithms to segregate cases into different partitions, Each partition is @ potential indicator of a subprocess or an archetype of a customer journey. Therefore, being able to partition the information ‘enables users to properly analyze more complex and bigger datasets. = According to our survey, EverFlow focuses more on business process improvement and audit and compliance use cases. Our survey revealed that it focuses significantly more than its competitors on real-time dashaoards with support for KPIs, followed by auditing and conformance checking capabilites. Fluxicon Based in Eindhoven, Netherlands, Fluxicon offers Disco version 2.14: + Disco is the most popular stand-alone, analysis-focused process mining tol on the market. Its visual appeal and usability are intuitive and understandable to business users, Therefore, no on-site professional services need to be delivered at the customer site, Disco is in the toolboxes of the igital professionals, and it can be used to quickly analyze al kinds of process probleme. + The Disco founders developed the first mining algorithm that was able to deal with complex processes, and they pioneered the “lider” approach for interactive process map simplification. As pioneers, the team at Fluxicon invested in the process mining community by ‘spreading the word and developing the methodology for applying process mining in practice. The compilation of this knowledge resulted in a ‘comprehensive process mining guide for every process mining practitioner. * ' Disco combines a solid scientific basis with practice, delivering some interesting capabilities such as process parallelism and analyzing the performance of a process while taking business days and working hours into account, + According to our survey, Fluxicon facuses more on business process improvement and audit and compliance use cases, Our survey revealed that it focuses significantly more than its competitors on modeling of processes and exceptions capabilites, followed by data preparation and data cleansing capabilities IBM Based in Armonk, New York, IBM provides IBM Process Mining version 1.12 = IBM entered the process mining market by acquiring mylnvenio. This fits into IBM's strategy to provide organizations with a one-stop shop of ‘Al-powered automation capabilities for business automation, including process mining, RPA, document processing, workflow and decisions. All these capabilities are tightly integrated with I9M Cloud Pak for Business Automation, IBM Cloud Pak for Integration, IBM Cloud Pak for Watson AlOps and IBM Cloud Pak for Network Automation + IBM provides an integrated BPA solution to analyze the process with a top-down approach within the wider context of derived processes from IBM Process Mining, Starting from this unified repository, IBM provides a dynamic model ofthe process that considers the wider ‘organizational context and leads to the representation ofa digital twin of an organization, 1 IBM Process Mining includes a simulation engine to analyze the Impact and ROI of any process change or to-be model before starting initiatives such as a digital transformation initiative. IBM Process Mining derives decision rules from the data and adds these rues into Decision Model and Notation (DMN) tables "= According to our survey, IBM Process Mining focuses more on improving process automation opportunities, and on the business process improvement use case. Our survey revealed thatit focuses significantly more than its competitors on predictive analytics capabilties, followed by intelligent support for enhancing process models Integris Based in Rome, Italy, Integrs offers Explora Process version 2.2: 1 lntegris is a product, consulting and system integration company that has added process mining to its data analytics and cognitive solution offerings. Integris completes its product offering with an advanced text mining framework that includes semantic and sentiment analysis, providing both automatic speech recognition and analysis. = Explora Process provides advanced filtering, process optimization simulations (Le., RA simulation), self-bullt reporting functionalities (using Microsoft embedded Power Bl) and time-based happy path and conformance evaluation, Those features allow time variant process: transformation that catches up powerful insights rather than frequency analysis. 1 Explora Process includes automatic log versioning, This allows full control over the analytic path, tracking every activity made by the analyst ‘over the log. Based on this fine-grained versioning tis possible to set up benchmarks between different versions of the same log with a refined graphical comparison,= According to our survey, Integrs focuses more on business process improvement opportunities and on audit and compliance use cases, Our survey revealed that it focuses significantly more than its competitors on support for realtime dashboards and KPIs, followed by process ‘modeling enhancement capabilities. Livejourney Based in Pars, France, Livejourney provides Livejourney version 3.9: ' Livejourney is one of the new entrants in the process mining market. After starting @ company that specialized in customer interaction ‘analysis, the founders realized thatthe optimizations made to these paths are also applicable to other processes: finance, logistics, industrial ‘and so on. This is how Livejourney, a French process mining solution, was born ' Livejourney has created a different user interface than most other available process mining offering, Livejourney provides a simple module (called Tags) that tags any fier in any part of the process discovery, process segmentation and process analysis ina visual way. By ‘connecting the tag toa specific user, te mechanism can be used in a collaborative way. '= By using predictive algorithms, Livejourney is able to predict next steps in the process and possible issues. Predicting the next bottlenecks in ‘a supply chain process, for example, makes it possible to increase or decrease the number of stocks or people needed to ensure the smooth running of the overall process. ‘= According to our survey, Livejourey focuses more on business process improvement and audit and compliance use cases. Our survey revealed thatt focuses significantly more than its competitors on data preparation and data cleansing capabilities, followed by realtime dashboards with support for KPls capabilities. Minit Based in Amsterdam, Netherlands, Minit provides Minit version 5.5: ' Minit, a process intelligence vendor, brings an easy-to-use, enterprisewide tool to the process mining market. Minit offers robust functionality ‘and collaboration around process performance dashboards, and It has been on the forefront of recognizing the importance and opportunities ‘of task mining and process simulations. = Minit analyzes complex processes based on different hierarchical levels, such as value-chain-process-subprocess mining, organizational structure mining, UI interaction recording (for RPA candidate identification) ane bot execution monitoring for hybrid business processes — ‘executed partly by human, partly by digital workforce. ‘= Minit added Business Rules monitoring that allows it te define, evaluate and notify the user about behavior ofthe processes (lke different process compliance issues). With the addition of Al-Powered Root Cause Analysis and Process Simulation, Minit provides a powerful toolset for continuous business process improvement, ‘= According to our survey, Minit focuses more on business process improvement and audit and compliance use cases. Our survey revealed that it focuses significantly more than its competitors on realtime dashboards with support for KPls capabiltes, followed by innovative ‘capabilities supporting task mining. Process Analytics Factory (PAF) Based in Darmstadt, Germany, Process Analytics Factory provides PAFnow 2021.6: = Process Analyties Factory was the first vendor in the process mining market to combine Bl software and services with process mining technology. The process mining component is part of its offering around process intelligence. = PAFnow is embedded in Microsoft Power Platform, and it extends Microsoft Power Bl with process mining capabilities, allowing Power BI Users to enhance ther current Bl infrastructures with process mining. Because Power Blis a key component of Microsoft Power Platform and Microsoft 365, users can use other Microsoft components, such as SharePoint, Teams, Power Automate for workflow automation and RPA, land Power Apps. "= PAFnovr is used mostly either on-premises (PAFnow in Power BI Report Server) or inthe cloud (PAF now in Power Bl service), andit is offered ‘as SaaS by Pracess Analytics Factory partners. PAF now allows clients to manage project tasks and progress in project management by tutning process mining insights into action in project management tools such as Asana, Tello or Jira Software According to our survey, Process Analytics Factory focuses more on improving aracess automation opportunities and on the audit and ‘compliance use case, Our survey revealed that t focuses significantly more than its competitors on execution capabilities, followed by data preparation and data cleansing capabilites.Puzzle Data Based in Seoul, South Korea, Puzzle Data provides ProDiscovery version 2.3: ‘= Puzzle Data is one of the frst Asian companies that developed a commercial process mining product. Closely connected to POSTECH, one of, the top universities in South Korea, Puzzle Data develops practical products using a balanced approach that satisfies customer needs, while ‘considering the latest theory. ‘= ProDiscovery introduces “puzzles” as kinds of widgets — small apps that run on a device's interface and provide extensibility, which is a feature that allows a snippet of an app to run within another app. Any artifact within process mining — such as process map, process pattern “analysis, social network analysis, workload analysis and case search — is represented as a puzzle. Each puzzle can be used in a dashboard, ‘and once fiters are applied to a dashboard, al the puzzles on that dashboard inherit those fiter conditions. + Puzzle Data has process mining implementations that are not trivial, such as long-lasting processes in a ship construction company, complex processes in high-tech manufacturing and fraud detection inthe online gaming industry, as well as processes for hospitals and third-party logistics providers. 1 According to our survey, Puzzle Data focuses more on business process improvement and digital transformation use cases, Our survey revealed that focuses significantly more than its competitors on support for customer interactions and customer journeys, followed by process modeling capabilities. QPR Software Based in Helsinki, Finland, QPR Software offers QPR ProcessAnalyzer version 2021.7: = QPR one of the frst offerings in the market, also provides a comprehensive set of performance management, BPM and EArelated software products and consulting services. The process mining offering is part of a more complete QPR Business Operating System offering, uses process mining analysis results within a digital twin of an organization. 1 QPRProcessAnalyzer extends the process mining concept with automated one-click root cause analysis, advanced influence and clustering ‘analyses, and machine learning and predictions, which identity case-tevel problems before they occur, ProcessAnalyzer can operate on top of, any data lake (e.g., Amazon Redshift or Snowflake) with its inmemory engine and single data policy. ‘= QPR has productized ready-made applications for six use cases (RPA, digital transformation, process improvement, process KPI reporting, IT ERP development, and auditing and compliance). These applications provide ready-made charts, dashboards and data source connections using the available QPR Connectors. ‘= According to our survey, QPR focuses more on the business process improvement use case and on improving process automation ‘opportunities. Our survey revealed that it focuses significantly more than its competitors on support for execution capabilities, followed by realtime dashboards with support for KPls capabilties. SAP (Signavio) Based in Berlin, Germany, Signavio provides SAP Signavio Process Intelligence version 16.3.0: ' ‘Signavio, a cloud vendor that specializes in innovative, easy-to-use products in the BPM/BPA and decision management markets, was ‘acquired by SAP in March 2021, Combining Signavio with the Business Process Intelligence unit of SAP will strengthen SAP's capacity to help companies quickly understand, improve, transform and manage their business processes at scale ‘= By blending process modeling, process mining and task mining (partnership with FortressIQ), SAP Signavio Process Intelligence accelerates process discovery, model creation, automation and transformation. Signavio explores the operationalization of customer excellence by mining internal business process operations and external customer journeys to identify inefficiencies and customer pain points across all customer touchpoints and journeys. = Based on its enterprisewide process platform and ready-toxse connectors, Signavio's objective is to democratize process mining, Its Process Collaboration Hub brings process mining insights to a broad audience and serves as a launchpad for more detailed process mining investigations. 1 According to our survey, Signavio focuses more on the business process improvement use case and on improving process automation ‘opportunites. Our survey revealed that it focuses significantly more than its competitors on capabilities supporting execution capabilites, {followed by support for customer interactions and customer journeys. Software AG Based in Darmstadt, Germany, Software AG delivers ARIS Process Mining 10.15:Software AG offers ARIS Process Mining as part of the ARIS suite, which is Software AG's offering fr business transformation and ‘operational excellence initiates. As one of the early commercial process mining offerings in the market, Software AG and global partners have implemented hundreds of process mining projects. 1 ARIS Process Mining does not depend on predefined process models in ARIS, but it can make use of the mined processes for process ‘conformance checking within the ARIS repository allowing ARIS customers to make the next step toward process maturity. 1 ARIS Process Mining does not stop at creating transparency and visisilty about process execution and uncovering bottlenecks and improvement potentials; it supports @ fully governed transformation of business processes, while involving all necessary stakeholders, ARIS ‘enables a fully controlled life cycle from pracess design, implementation and execution to process mining, improvement and transformation in one integrated sult. According to our survey, Software AG focuses more on business process improvement and audit and compliance use cases, Our survey revealed that it focuses significantly more than its competitors on process modeling capabilities, followed by capabilities supporting ‘conformance checking, gap analysis and continuous monitoring, Soroco Based in Boston, Massachusetts, and Bangalore India, Soroco provides Scout Platform version 4.6 + Soroco is one of the new entrants in the process mining market. The company intially aunched Scout to focus on improving tasks by understanding the "work graphs" of how people within companies worked digitally. Later on, it expanded this task mining with process mining, bbased on the same work graph modeling techniques. = The Scout platform combines the latest process mining and task mining within its overall process discovery capabilities and coins this as “hybrid process mining” Scouts “deep capture" technology provides in-depth and advanced data-driven insights across multiple lenses, ‘Subsequently, Scout provides actionable insights about how to enhance the process by tll reduction, firstly from a process improvement or ‘excellence perspective, then from an automation perspective. 1 Scout captures click and keystroke interactions between the end user and their desktop/laptop directly inthe operating system, which it calls, “deep capture,’ This capture capability alleviates the requirements for various integration challenges, allowing projects to start quickly, ‘= According to our survey, Soroco focuses more on business process improvement and support for process automation use cases. Our survey revealed that it focuses significantly more than its competitors on process model enhancement, followed by task mining, inferring useful information from lowlevel event data available in UI logs. StereoLOGIC Based in Toronto, Canada, StereoL OGIC provides Stereol OGIC Process and Task Mining 2021 ' ‘StereoLOGIC, one of the fist offerings in the market, is a company that has, from the ground up, developed its own mechanisms and algorithms for process discovery, analysis and conformance checking 1 StereoLOGIC captures system-based events, as well as user interface-based information, without the need to install anything on the employee workstation. It recognizes that most employees perform tasks belonging to several processes, and these tasks interact many times. ‘StereoLOGIC offers 2 more complete picture ofthe reality and can handle employee productivity comparisons of workload optimiz different processes and related tasks. This allows StereoL OGIC to capture the detalled interaction of employees with productivity applications. 1 ‘StereoLOGIC is installed on-premises or ina private or external cloud. StereoLOGIC monitors work across multiple levels, starting from business activity monitoring down to task mining, process mining, processes discovery and precise diagnostics in realtime. ' According to our survey, StereoLOGIC focuses more on the business process improvement use case and on improving process automation ‘opportunities. Our survey revealed that It focuses significantly more than its competitrs on task mining capabilities, followed by capabilties ‘supporting the interaction of different processes, UiPath Based in New York, UiPath provides UiPath Process Mining and UiPath Task Mining version 2021.10: 1 1n 2019, UPath acquired ProcessGold, a process mining company, to complement its market-leading RPA platform with capabilities to discover opportunities for RPA implementations and to contextualize and monitor the implemented bots, Furthermore, the process mining ‘component delivers an essential piece of the hyperautomation strategy of UiPath,‘= Inheriting the platform thinking and the multidimensional process mining approach of ProcessGold, makes it possible for UiPath Process Mining and Task Mining to look at all attributes on two or more related processes and tasks in a single model. The analysis of these multiple interactions ereates new insights and enriches the UiPath hyperautomation platform, '= The integration with Automation Hub enables UiPath Process Mining users to exaort automation ideas surfaced from UiPath Process Mining to Automation Hub to speed up implementation time and avold manual information transfer, TemplateOne isa fully customizable, flexible and reusable starting point to develop custom UiPath Process Mining dashboards and KPIs based on client needs, ‘= According to our survey, UiPath focuses more on business process improvement and on aucit and compliance use cases. Qur survey revealed that it focuses significantly more than its competitors on realtime dashoards with support for KPls capabilities, followed by capabilities ‘supporting process models, Market Recommendations To drive business transformation and optimization through technology innovation, business executives and data and analytics leaders should: ‘= Improve visibility and understanding ofthe actual performance of business operations and processes by investing in process mining. Actual quantitative data is delivered in a context that nat only reveals information about a process, but also connects this data to other constituents in a value chain, such as data about clients, products, services, channels, resources, applications and roles. This richness of information will improve the business outcomes of any automation initiative, whether at a task, workplace or process level ' Create awareness and inspire business and operational colleagues on the benefits, capabilities and use of process mining by introducing ‘small, short-term pilots. To mine a process, not al the activities and their related data should be captured at once. Start a pilot on activities Where the data is easily available. This starter project wil deliver value and will provide insights in where the next iteration needs more detailed data, This iterative or agile approach will create valuable insights and fuel the acceptance, enthusiasm and demand for larger process mining projects. ' Explore use cases that go beyond traditional, internally focused process discovery, process validation and model enhancement by targeting business operations and interactions with external parties such as customers. This can be seen as customer journey mining — opening up a new audience and aligning customer journey maps with internal process maps. 1 Assess the required scalability of the process mining initiative by defining the intended users ofthe process mining results and by defining the volume, frequency patter and concurrency of operational data to be mined. This will helo in tool selection because not all vendors target the same sealailly Evidence Excellent research and books have been published on the topic of process mining, Some ofthese extensive materials include: 1 Process Mining: Data Science in Action, Coursera ‘= W-van der Aalst, “Process Mining: Data Science in Action Springer Verlag, 2016, Process Mining: A Database of Applications, HSPI Process Mining + Process Mining Manifesto, IEEE Task Force on Process Mining 2 i. van der Aalst, “Process Mining: Data Science in Action, Springer Verlag, 2016, ® Business operations model: A business operations model is the operationalization of the business model through the identification, representation and measurement of the business model constituents, as well as the interactions between these components to deliver the expected stakeholder value, Examples include detailed value chains; demand-driven value networks; dynamic capability models; and enterprise ‘supplier, input, process, output and customer (SIPOC) models. “The founders have compiled their knowledge into a comprehensive process mining guide for every process mining practitioner. Note 1: Representative Vendor Selection Gartner sizes the market at about 40 vendors, which offer most of the process mining capabilities. For the 20 vendors profiled here, Gartner Used client interactions, vendor briefings and academic research to verify that these offerings were in the market for at least a year and that they represent examples ofthis market, In terms of revenue, we estimate that these vendors represent more than 90% of the total marketNote 2: Gartner's Initial Market Coverage This Market Guide provides Gartner intial coverage ofthe market and focuses on the market definition, rationale for the market and market, dynamics. Note 3: Interview With Prof.dr.ir. Wil van der Aalst, Lehrstuhl fiir Informatik 9/Process and Data Science, RWTH Aachen University From October 2021 What are the main changes in the process mining space? “The adoption of process mining continues to increase. In countries lke Germany and the Netherlands, most ofthe larger companies are already Using process mining. Each year the process mining market increases by 50% or more. There are currently around 40 process mining vendors, [At the same time, there were some mergers. For example, SAP acquired Signavio, ]M acquired mylnvenio, and Appian acquired Lana Labs. This, shows that many of the lager software vendors now see the importance of process mining. With process mining being integrated ito diferent platforms, the threshold to use process mining gets much lower. For example, there are at least five process mining tools that build on Power BL In terms of process mining capabilities, the focus is shifting from discovering simple Directly Follows Graphs (DFGs) to more advanced capabilites ike conformance checking prediction, and simulation using richer languages lke the Business Process Model and Notation (BPMN). An important development is also the integration of process mining and automation platforms. By using process mining, execution ‘gaps are identifies while automatically triggering corrective workflows. This is natural development. Process mining provides the X-rays to spot performance and compliance problems, However, after diagnosing these problems, one also needs to “operate — ie, take the actions to address the execution gaps uncovered. Therefore, process modeling, process analysis, and process execution need to be integrated. The attention for topes lke hyperautomation and execution management ilustrates this.” What are examples of new process mining capabilities that will become important in the next couple of years? “Although process mining vendors already offer machine learning (ML) capabilties and often refer to artificial intelligence (A\), the actual practical useis very limited. This has two reasons. The first reason is that in many organizations, there is s0 much collective debt that ML and {Al approaches are not needed and also do not work. As long as organizations have data integration problems, they should not start with ML/AL Traditional process mining approaches lke process discovery and conformance checking will eveal many improvement opportunities. Sophisticated forward-looking process mining techniques (e.g, using ML to predict delays and deviations) only make sense after taking care of the obvious. Moreover, these techniques only workif enough data and processes are in steady-state. | often advocate hybrid inteligence (HI) as the more realistic alternative. Hence, Im skeptical about fully automated ML/Al. For me, a much more exciting development is the emergence of object-centrc process mining (OCPM), Ths is more down-to-earth, but very Important for future process mining applications. Most process mining tools still assume that there isa single case identifier. However, one event may refer to multiple objects (e.g, an order, a customer, multiple items, and a location). Picking a particular case nation quickly leads to convergence and divergence probleme, and the connection between abjects gets lost, Take, for example, alook atthe database tables used to ‘analyze purchase-o-pay (P2P) or order-to-cash (020) processes. These tables are connected through foreign keys showing that one order may have multiple items and that a delivery may contain items of multiple orders. As the field matures, OCPM will become more important. Organizations do not want to spend efforts on the repeated extraction of events to answer specific questions.” How did the COVID-19 pandemic influence the adoption of process mining? “Due to the COVID-19 pandemic, many organizations cut budgets for innovation, and people were busy keesing the daly operational processes Lp and running, However, COVID-19 was also an enabler of diitalization initiatives, Many decisions that would normally take years took place in ‘a matter of weeks — e.g, replacing paper forms with electronic forms. The pandemic exposed inefficiencies and outdated work practices. In the long run, these experiences will bea catalyst for process mining adoption. The pandemic also showed the limitations of mainstream ML/Al technologies, In times of disruption, “raining data” are not representative of the actual behavior of people and organizations. However, process ‘mining can stil provide valuable insights when things change dramatically. Moreover, the resilience of processes will be atopic in the boardrooms of most organizations.” How many process mining vendors will tere be in 2025? “The number of process mining vendors has been steadily growing. Now there are around 40 commercial process mining tools. Some of the current vendors will be acquired by larger vendors and their process mining capabilties will be added to existing platforms and systems. Basic process mining capabilites (eg, creating a DFG with frequency and time information) will be added to existing systems (e.g, RPA and BPM software). Also new, domain-specific vendors (e.g, targeting production or healthcare) will emerge. However, it will be increasingly difficult to compete withthe leading vendors ofthe highrend process mining tools that are able to deal with billions of events. Hence, itis dificult 0 estimate the exact number of process mining vendors in 2025, However, itis clear that the market wll continue to grow, and the diversity of tools supporting process mining will increase”can you tell more about hybrid intelligence? “Hybrid inteligence (HI), sometimes also called augmented intelligence, emphasizes the assistive role of new data-driven technologies. For ‘example, deep neural nets are there to enhance human inteligence rather than to replace it (just Ike telescopes are there to enhance human Vision). HI aims to combine human and artificial intelligence, thereby reaching superior results to those each of them could have accomplished ‘separately while learning continuously. HI aims to combine the best of both worlds, The spectacular developments in ML have extended the reach of software and hardware robots. Once a robot is able to perform a repetitive task at a similar level of quality, itis often also more cost- effective. However, humans sil have unique capabiliies. Consider, for example, disruptive events lke the COVID-19 pandemic, where one is. confronted with completely new challenges that require flexibility, creativity and intuition, People have the ability to transfer experiences from ‘one problem domain to another. Moreover, empathy (ie. the capacity to understand or feel what another person is experiencing) and ethics (ie. reasoning about moral concepts such a good and evil right and wrong, virtue and vice, justice and crlme) require Hl HI also plays a key role in action-oriented process mining (AOPM), which focuses on automated corrective actions based on process mining diagnostics. AOPM tums observed events into management actions when needed, The goal is not fo support the operational process itself (that already exists in some form), but to support the management of the process. Obviously, Hl will play a significant role in AOPM, Humans need to be in the loop when confronted with unprecedented situations. For example, it may be beneficial to be noncompliant when novel problems ‘emerge. Ths is what we have learned in the last two years." Learn how Gartner canhelp you succeed (© 2022 Garner, ne. andor its ffilaes Allrights reserved. Gartner isa registered trademark of Garner, nc and its affllates. This publication may not be reproduced or citriouted in any form without Gartner prior wren permiszion, I consist of the opinions of Carters research organization, which should not be construed as statements of fact While the information contained inthis publication has been obtained from sources beleved tobe reliable Gartner disclaims all warranties as tothe accuracy, completeness or adequacy of such information, Although Gartner research may address legal and financial issues, Gartner does not provide legal or investment acvice and its research should nat be consti or used as such. Your access and use ofthis publication are governed by Gartner's Usage Poly, Gartner prides itself on ts reputation for indoperdenes and abjctvy Rs research is produced independently by ts esearch organization without Input or in ence trom any third party, For further information, eee "Guiding Principles on Independence and Objectivity. About Careers Newsroom Policies Site Index IT Glossary Gartner Blog Network Contact Send Feedback