Context. This thesis pertains to the field of software architecture evolution, focusing on the migration of legacy monolithic applications to microservice-based architectures using a tool-supported, quality-driven method, in particular the Architecture Refactoring Helper [Fri24a]. Problem. Monolithic applications that grow large often decrease developer productivity [VGC+15]. Organizations that want to migrate their monoliths towards microservices previously had no guidance to find a Service Identification Approach that aligns with their demands [Her24; Kno23]3.3. Objective. This thesis, wants to validate the applicability of the Architecture Refactoring Helper [Fri24a] in constrained real-world environments, through a collaborative case study with the LF Consult GmbH [LF 24], by performing a proof of concept decomposition on a subcomponent of the 3LiterPPS-Server®. Additionally, this thesis suggests improvements for the shortcomings found. Method. During this thesis, we conducted a scenario-based architecture review with developers and the software architect, as well as a preference review with LF Consult GmbH’s [LF 24] Head of Development and Support, to configure the Architecture Refactoring Helper [Fri24a]. Afterward, we used different configurations to query for Service Identification Approaches. After attaining the highest-ranked applicable SIAs, we performed the proof-of-concept decomposition, compared, and evaluated the candidate services. Lastly, we evaluated the Architecture Refactoring Helper, suggested improvements, and discussed threats to validity. Result. The ARH’s [Fri24a] phased methodology systematically narrowed over one hundred SIAs to the Service Cutter [GKGZ16a] approach, yielding reproducible service candidates for the Dispo component in the 3LiterPPS-Server®, while revealing limitations in match aggregation, preference weighting, and UI clarity. Conclusion. The ARH [Fri24a] proves valuable for structured migration planning under realistic constraints, though enhancements in metric expressiveness, preference exclusion, and weighting, and interface usability are recommended to strengthen decision support.