A key objective of knowledge management system (KMS) implementations is to facilitate job outcomes, such as job performance and job satisfaction. Prior KMS research indicates many KMS implementations have failed to achieve their intended job outcomes, such as job performance and job satisfaction, and one important reason for failure could be that employees do not know how to use a KMS to enhance job outcomes. Given that research on this topic is scant and the findings inconsistent, this paper sought to develop a better understanding of the topic. Specifically, we examine how employees can use a small number of KMS features to get a majority of their job tasks done. Limited research has used a systematic approach to identify these features, examined drivers of using these features, and impacts of the use of such features on job outcomes. Based on a literature review, we first identified several KMS features. Then, these features were examined using a qualitative study among 35 employees in a large organization in the finance industry to identify the key KMS features that could contribute positively to job outcomes. We then developed a nomological network of KMS feature use. Leveraging social network theory, we present peer support ties in general, and help-seeking ties and help-providing ties in particular, as key drivers of the use of these features and job outcomes. We also present various competing hypotheses for the effects from peer support to KMS feature use, KMS feature use to job outcomes, and peer support to job outcomes. We conducted a quantitative study (n = 1,441) in the same organization (noted above) to validate our model. Results indicated that our model was largely supported.