The EU Innovation Union identified that innovations are needed to boost economic growth performance, where SMEs are considered vital for these goals. To achieve this, business model innovation (BMI) is identified as the new area to innovate and also source of future competitive advantage. Furthermore, business model changes as a form of innovation can be seen as a response to the changes of external environment or internal factors of the firm. One of the barriers in changing business model would be the conflict between existing BM and its underlying operation with the new BM. Moreover, by taking account the external and internal factors of the firms, there can be various arrangements in the BM and operational elements that can be modified by the firm, making it a complex and cumbersome process. To overcome these barriers, ontologies or frameworks could be utilized to guide the BMI practice. However, firms are presented with various options of ontologies with different possible changes in both BM and operational elements. Furthermore, while trial-and-error approach could be utilized, firms don't have much time to experiment due to market pressure. Hence, this research found a need to provide SMEs with insights regarding pattern of past BMI practices to help simplify and reduce the time to perform BMI. To achieve this, this research took several steps. First, to have an aggregated learning regarding BMI patterns with a limited time and resources, the author selected the case survey method as the research approach. Case survey method are used because it can provide inexpensive way of tapping the rich insights from various cases in a relatively shorter time than performing individual case studies. Second, the case survey coding scheme was designed by exploring vast amount of literature. Due to the exploratory nature of this research, the case survey coding scheme are made to have open-ended questions. Third step would be the data collection. To collect the BMI cases, the author contacted several researchers in Europe and members of ENVISION project that have done case studies in business model innovation to have a more relevant focus and time-saving (purposive sampling). Fourth, the case survey coding scheme was applied to the BMI cases. To help the coder in assessing the BMI cases, this research included a coding manual that consists of definitions and rules regarding each variables of the coding scheme. Fifth, this research converted the qualitative data into quantitative data by using qualitative content analysis. Quantitative coding was done by counting the frequencies of each answers categories across all cases. Sixth, due to the interpretive nature of the coding, this research used alternative means to measure reliability and validity of the qualitative coding. Reliability of the coding are being measured through dependability (e.g. transparent coding process / coding manual) and confirmability (e.g. member checking). There are several findings regarding the BMI practice done by SMEs. First, the major drivers for SMEs to do BMI would be due to market dynamics, high innovativeness and low business performances. Second, the most BM ontologies used to guide the BMI would be Canvas and STOF, while ArchiMate is found to be the most popular EA framework used. Third, we found that in overall, changes in the BM are mostly related to changes in services, organizational network and target market. And lastly, the changes in BM will create changes in the operational area especially in the process domain, value dependencies and information domain. This research contain many limitations, and the BMI patterns in this research are mainly descriptive insights, which can be improved by using cluster analysis, configuration analysis and network analysis.