A family-run small Swiss business that operates in the specialized construction activities sector has utilized its software solution for approximately 20 years. Recently, the current software solution failed to meet the requirements of the evolving small business. Factors such as duplicate data entries, media discontinuities, and missing analysis capabilities disrupt daily operations. In addition, a foreign investor acquired the software vendor. Hence, this thesis systematically develops a decision basis for the procurement and recommendations for the implementation of the new solution.

The thesis applied a single case for the study, in which twelve core processes were documented as use cases. Based on 157 artifacts, 12 observation protocols, and 16 semi-structured interviews, 99 requirements were elicited. The requirements were also classified with help of the Kano model and prioritized by their necessity according to the MoSCoW model. The evaluation applied a two-tier preselection. The exclusion criteria decimated the field of possible vendors. The remaining candidates were subject to a utility analysis, including a subsequent sensitivity test for robustness of the result.

The requirements catalog reflects the expectations of the company and illustrates four recurring themes. Single data entry with subsequent reuse, mobile data entry at the service location, automation of administrative tasks, and transparency of work statuses. 15 vendors were examined based on an industry association's software partner list. In the final utility analysis, one vendor ranked first not only in the base case but also in all sensitivity runs. Although the first rank is robust, the margin between the two vendors reaching the utility analysis depends on the treatment of criteria for which only partial evidence was found. Hence, this thesis recommends the procurement of the software ranking first based on the premise that unanswered criteria can be verified during vendor demonstrations. The implementation is supported by six recommendations, among others: the definition of the target processes, a lean organization of the implementation project, configuration of the software within its standard, and acceptance tests based on the 12 core processes. The procedure of this study can be transferred to other comparable small companies that face similar software decisions.