Faced with the challenges to addressing the novelties of the changing business environments (e.g., new customer requirement, changes in customers taste and preferences, the introduction of new products or services by competitors), organizations seek to build collaboration among their employees who possess complementary knowledge. Integrating complementary knowledge enhances employees' ability to address environmental challenges and foster innovation. Despite the importance of knowledge integration for innovation, integration of such knowledge becomes difficult when employees lack a shared understanding of knowledge, and when the knowledge is newly generated. Because new knowledge is tacit in nature and highly personal to a particular individual, it is difficult to articulate, making knowledge integration (KI) an arduous task. Lack of shared understanding, the presence of new knowledge, and lack of common interests in employees creates three types of knowledge boundaries – syntactic (information processing) boundaries, semantic (interpretive) boundaries, and pragmatic (political) boundaries. The presence of knowledge boundaries makes it difficult for employees to share and access their knowledge with each other. To overcome the challenges related to the knowledge boundaries, employees use boundary-spanning objects, which are common lexicons, common meaning, and common interests, to share and access their knowledge across the boundaries. Although prior studies have emphasized the importance of knowledge integration of various knowledge sources for innovations, examinations of what enhances KI capability of employees for organizational innovation remain limited. In addition, apart from Carlile, (2004) and Franco (2013), which are both case studies, other studies that examine the role of boundary spanning objects for knowledge integration are missing. The knowledge management literature also fails to measures (the success of common lexicons, common meaning, and common interests for achieving KI capability) boundary spanning objects. Therefore, in this study, new measurement items of boundary spanning objects and novelty are developed to test the hypotheses. A survey-based design was used to collect data and measure the constructs examined in this study. Structural equation modeling (SEM) was used to test the direct relationship hypotheses. The moderation effects were tested using 1) multi-group analysis using hierarchical linear regression, and 2) relative weight of each boundary spanning object determining KI capability at the different levels of novelty. Evidence suggests that while common meaning and common interests positively influence KI capability, common lexicon does not have a statistically significant relationship with KI capability. The results also revealed that KI capability positively influences organizational innovation. Moreover, the results demonstrate that the strength of the relationship between boundary spanning objects and KI capability is different at the medium and the high level of novelty.