Abstract
As the service economy grows, service providers' success increasingly hinges on delivering customer-centric services. The customer journey approach, which maps service delivery from the customer's perspective, has gained prominence recently. However, despite its widespread adoption, there's a notable absence of standardized methods and formalisms in customer journey mapping. The Customer Journey Modelling Language (CJML) presents a user-friendly visual language that includes specific terminology, syntax, diagrams, methods, and tools. Nonetheless, the existing tools for creating CJML visualizations are significantly limited. This thesis aims to develop a tool for diagramming and analysis of customer journeys based on CJML. The thesis raises the research questions such as what features are implemented in other tools, and what design features and principles could enhance visualisation and analysis and usefulness of the tool. The created tool is called “CJML analyser” and relies on xCJML, a data format tailored for precise specification of both hypothetical and real customer journeys. The CJML analyser presents three methods of interaction: A drawing canvas, input via a structured form, and import of xCJML files. Notable features include diagram switching, validation of actual journey based on planned journey (conformance), and basic statistics. The CJML analyser was evaluated through a usability test with eight target users. Here, the test participants completed three predefined tasks using the tool. After completing the tasks, the participants provided feedback, and filled out the "System Usability Scale" questionnaire. The CJML analyser was well received by the participants with an average SUS score of 72 (out of 100). However, the evaluation also revealed several points of improvement, in particular concerning clarity and design of the input methods. Limitations of the study include the small sample size and the reliance on unstructured feedback. Future work for the CJML analyser should simplify the form input method, interactions with the “undo and redo” functionality, large language model “ChatGPT” implementation into the tool for in-depth analysis, and low threshold data input from “Microsoft Excel” templates.