How a BPM Dream Team Ranked the Risks and Tools for AB-BPM
To address the research questions, the grounded theory (GT) research method was employed, in combination with the Delphi method. GT is suitable for developing theories in areas with limited prior research. This approach was chosen due to the lack of…
To address the research questions, the grounded theory (GT) research method was employed, in combination with the Delphi method. GT is suitable for developing theories in areas with limited prior research. This approach was chosen due to the lack of existing studies on the industry's perspective on applying AB testing for Business Process Improvement (BPI). Initial data collection was conducted through semi-structured interviews, followed by a second stage of theoretical sampling, adopting a shortened Delphi method study. The ranking-type Delphi method (RTDM) was utilized to identify and rank key issues related to the topic.
Experts were recruited from a multinational software company with over 100,000 employees, primarily from a sub-unit specializing in Business Process Management (BPM) software. The selection criteria included individuals involved in BPM software development and consultancy, covering various technical skills such as software engineering and data science. The panel consisted of ten experts, with participants averaging 7.6 years of industry experience and 4.3 years in BPM. Educational backgrounds varied, with degrees in Science/Engineering and Business/Management. The study involved three rounds: interviews, a validation survey, and a ranking survey.
Semi-structured qualitative interviews were conducted to explore a wide range of ideas and thoughts on the topic. Interview guidelines covered prior experience with BPI, a brief introduction to AB-BPM, execution feasibility, suitability, prerequisites, risks, tool requirements, and open discussion. Adjustments to the guide were made during interviews to reflect newly gained insights.
Post-interview, transcripts were coded, and topics were consolidated. Categories such as risks and tool features were selected for further data collection due to high expert interest. Experts validated the representation of their stances and provided feedback on any missing items. The narrowing-down phase was omitted due to the short initial lists, aligning with standard practices.
To address the research questions, the grounded theory (GT) research method was employed, in combination with the Delphi method.
The ranking phase involved evaluating the importance of issues, with separate metrics for risks and tool features. Risks were ranked based on the perceived likelihood of occurrence and potential damage, resulting in scores from 1 to 25. Tool features were rated on a Likert scale from extremely unimportant to extremely important. Multiple ranking rounds were not conducted, focusing instead on exploring new insights.
Aaron Friedrich Kurz, Technische Universität Berlin and SAP Signavio, Berlin, Germany Timotheus Kampik, SAP Signavio, Berlin, Germany Luise Pufahl, Technische Universität München, Munich, Germany Ingo Weber, Technische Universität München and Fraunhofer Gesellschaft, Munich, Germany
This paper is available on arXiv under the CC BY 4.0 DEED license.
Based on reporting by hackernoon.com.