ECIS 2026 (2026) AI Processed
UNPACKING THE ROLE OF INDIVIDUALS IN PROCESS MINING: A LITERATURE REVIEW AND RESEARCH FRAMEWORK
This study conducts a systematic literature review of 23 academic articles to consolidate and analyze research on the individual level of process mining. The authors synthesize the fragmented literature to map out how individual behaviors, usage patterns, and emotional affect interact with various contingency factors to drive business value.
Problem
Prior process mining research has predominantly focused on technical algorithms or organizational-level adoption, leaving the critical role of individual users largely overlooked. This gap results in fragmented terminology and lack of a coherent understanding of how individual actions and decision-making actually translate into realized business value.
Outcome
- The review identifies and establishes user behavior, usage patterns, and affect (such as uncertainty and trust) as the three core elements constituting the individual level of process mining, supported by the synthesized literature.
- The analysis reveals that technical, individual, and organizational contingency factors systematically influence user behavior and affect; for example, data inaccuracies undermine user trust, while formal training and methodological guidance support correct execution.
- Individual behaviors and affective states are shown to directly impact value outcomes, with structured behaviors like hypothesis testing decreasing analysis duration and increasing the quality of insights.
- Key limitations of the study include its reliance on a limited sample of 23 reviewed papers, a static framework that does not capture sequential or temporal dynamics, and a lack of true inter-coder reliability during the qualitative coding process.
- The analysis reveals that technical, individual, and organizational contingency factors systematically influence user behavior and affect; for example, data inaccuracies undermine user trust, while formal training and methodological guidance support correct execution.
- Individual behaviors and affective states are shown to directly impact value outcomes, with structured behaviors like hypothesis testing decreasing analysis duration and increasing the quality of insights.
- Key limitations of the study include its reliance on a limited sample of 23 reviewed papers, a static framework that does not capture sequential or temporal dynamics, and a lack of true inter-coder reliability during the qualitative coding process.
What it means for you
- CIO / IT Executive: Initiate a data audit of the source pipelines feeding your enterprise process mining tools to eliminate data discrepancies, directly addressing the research finding that data inaccuracies severely undermine user trust and derail tool adoption.
- IT Manager: Add a mandatory 'Hypothesis-First' template to your team's process mining workflow on Monday morning, requiring analysts to document their specific diagnostic hypothesis before running queries to enforce structured behavior and reduce analysis duration.
- Business Strategist: Redesign the process mining ROI dashboard to measure actual business value outcomes (e.g., cost savings from process changes) rather than user log-in frequency, linking individual analytical insights directly to operational KPIs.
- Researcher: Draft a research design for a longitudinal study tracking 20 process mining analysts over six months to capture the sequential and temporal dynamics of user affect and trust, directly addressing the static limitations of current literature.
- Policymaker: Establish an internal organizational standard mandating that no department may deploy process mining software without integrating a formal training and methodological guidance program, ensuring compliant and high-quality data execution.
Transcript
Host: Hello and welcome to A.I.S. Insights - Turning IS Research into Business Action. I'm your host, Anna Ivy Summers. Today, we are exploring a study presented at the European Conference on Information Systems, or ECIS 2026, titled "UNPACKING THE ROLE OF INDIVIDUALS IN PROCESS MINING: A LITERATURE REVIEW AND RESEARCH FRAMEWORK". Joining me to discuss this is our resident expert, Alex Ian Sutherland. Welcome, Alex.
Expert: Thank you, Anna. It is great to be here. This study conducts a systematic literature review of twenty-three academic articles to consolidate and analyze research on the individual level of process mining. The authors synthesize what has been a fragmented literature to map out how individual behaviors, usage patterns, and emotional affect interact with various contingency factors to drive business value.
Host: Let us start with the problem this study addresses. Process mining has been around for some time, so why is focusing on the individual user so important now?
Expert: Historically, process mining research and practical implementations have focused heavily on either the technical side, like developing new algorithms, or the organizational side, like corporate adoption and governance. However, this has left a significant gap regarding the actual individual users who engage with these tools. These individuals are the ones who must interpret the data, make sense of process models, and decide on improvement actions. Because this individual level has been largely overlooked, we have faced fragmented terminology and a lack of a coherent framework explaining how individual actions and decision-making actually translate into realized business value.
Host: How did the authors of this study approach this problem?
Expert: They conducted a representative literature review following established guidelines in the Information Systems field. After querying databases and applying strict inclusion criteria, such as ensuring the research went beyond simple software usability evaluations, they analyzed a final sample of twenty-three peer-reviewed articles. They then utilized qualitative coding to synthesize these studies and build a cohesive research framework.
Host: What did they find when they analyzed these twenty-three studies?
Expert: First, they established that three core elements constitute the individual level of process mining: user behavior, usage patterns, and user affect, which refers to states like trust and uncertainty. Second, they found that various technical, individual, and organizational contingency factors systematically influence these three elements. For instance, on a technical level, data inaccuracies directly undermine user trust in the system. On an individual or organizational level, formal training and methodological guidance support correct execution and help prevent errors.
Host: And how do these individual behaviors actually impact the business value of process mining?
Expert: The study shows that structured behaviors have a direct, positive impact on outcomes. For example, when analysts actively create and test hypotheses during their work, it decreases the duration of the analysis and increases the overall quality of the insights. On the flip side, when users face difficulties or use incorrect approaches, it can lead to misinterpretations and poor-quality results.
Host: Were there any limitations mentioned in the study that we should keep in mind?
Expert: Yes, the authors noted several limitations. First, the framework is based on a relatively small sample of twenty-three reviewed papers. Second, it is a static framework, meaning it does not fully capture the sequential or temporal dynamics of how users work over time. Finally, the qualitative coding process lacked true inter-coder reliability, as the initial coding was conducted primarily by the first author, though it was discussed in weekly meetings with the co-authors.
Host: That is very helpful context. Let us talk about the practical side. How can business and technology leaders apply this knowledge in their organizations?
Expert: Practitioners can take several concrete actions based on these findings. First, since data quality directly impacts user trust, organizations must invest in data preprocessing and validation before giving analysts access to tools. If analysts do not trust the data, they will be reluctant to use the tools effectively. Second, because structured behaviors like hypothesis testing improve insight quality, training programs should move beyond just teaching software features. They should train users in structured analysis methodologies. Finally, managers should provide clear methodological guidance and support, such as templates or standard procedures, to reduce user uncertainty and prevent common interpretation mistakes.
Host: So, to summarize, process mining does not create value on its own; it depends heavily on the individual's behavior, trust, and training. By focusing on data quality, structured training, and methodological support, businesses can significantly improve the speed and quality of their process insights.
Expert: Exactly, Anna. It is about moving from a purely tool-centric view to a human-centric view of process mining.
Host: Thank you, Alex, for sharing these insights, and thank you to our listeners for tuning in to A.I.S. Insights. We will see you next time.
Expert: Thank you, Anna. It is great to be here. This study conducts a systematic literature review of twenty-three academic articles to consolidate and analyze research on the individual level of process mining. The authors synthesize what has been a fragmented literature to map out how individual behaviors, usage patterns, and emotional affect interact with various contingency factors to drive business value.
Host: Let us start with the problem this study addresses. Process mining has been around for some time, so why is focusing on the individual user so important now?
Expert: Historically, process mining research and practical implementations have focused heavily on either the technical side, like developing new algorithms, or the organizational side, like corporate adoption and governance. However, this has left a significant gap regarding the actual individual users who engage with these tools. These individuals are the ones who must interpret the data, make sense of process models, and decide on improvement actions. Because this individual level has been largely overlooked, we have faced fragmented terminology and a lack of a coherent framework explaining how individual actions and decision-making actually translate into realized business value.
Host: How did the authors of this study approach this problem?
Expert: They conducted a representative literature review following established guidelines in the Information Systems field. After querying databases and applying strict inclusion criteria, such as ensuring the research went beyond simple software usability evaluations, they analyzed a final sample of twenty-three peer-reviewed articles. They then utilized qualitative coding to synthesize these studies and build a cohesive research framework.
Host: What did they find when they analyzed these twenty-three studies?
Expert: First, they established that three core elements constitute the individual level of process mining: user behavior, usage patterns, and user affect, which refers to states like trust and uncertainty. Second, they found that various technical, individual, and organizational contingency factors systematically influence these three elements. For instance, on a technical level, data inaccuracies directly undermine user trust in the system. On an individual or organizational level, formal training and methodological guidance support correct execution and help prevent errors.
Host: And how do these individual behaviors actually impact the business value of process mining?
Expert: The study shows that structured behaviors have a direct, positive impact on outcomes. For example, when analysts actively create and test hypotheses during their work, it decreases the duration of the analysis and increases the overall quality of the insights. On the flip side, when users face difficulties or use incorrect approaches, it can lead to misinterpretations and poor-quality results.
Host: Were there any limitations mentioned in the study that we should keep in mind?
Expert: Yes, the authors noted several limitations. First, the framework is based on a relatively small sample of twenty-three reviewed papers. Second, it is a static framework, meaning it does not fully capture the sequential or temporal dynamics of how users work over time. Finally, the qualitative coding process lacked true inter-coder reliability, as the initial coding was conducted primarily by the first author, though it was discussed in weekly meetings with the co-authors.
Host: That is very helpful context. Let us talk about the practical side. How can business and technology leaders apply this knowledge in their organizations?
Expert: Practitioners can take several concrete actions based on these findings. First, since data quality directly impacts user trust, organizations must invest in data preprocessing and validation before giving analysts access to tools. If analysts do not trust the data, they will be reluctant to use the tools effectively. Second, because structured behaviors like hypothesis testing improve insight quality, training programs should move beyond just teaching software features. They should train users in structured analysis methodologies. Finally, managers should provide clear methodological guidance and support, such as templates or standard procedures, to reduce user uncertainty and prevent common interpretation mistakes.
Host: So, to summarize, process mining does not create value on its own; it depends heavily on the individual's behavior, trust, and training. By focusing on data quality, structured training, and methodological support, businesses can significantly improve the speed and quality of their process insights.
Expert: Exactly, Anna. It is about moving from a purely tool-centric view to a human-centric view of process mining.
Host: Thank you, Alex, for sharing these insights, and thank you to our listeners for tuning in to A.I.S. Insights. We will see you next time.