In organizations, individuals typically rely on their personal networks to obtain expertise when faced with ill-defined problems that require answers that are beyond the scope of their own knowledge. However, individuals cannot always get the needed expertise from their local colleagues. This issue is particularly acute for members in large geographically dispersed organizations since it is difficult to know ?who knows what? among numerous colleagues. The proliferation of social computing technologies such as blogs, online forums, social tags and bookmarks, and social network connection information have expanded the reach and ease at which knowledge workers may become aware of others? expertise. While all these technologies facilitate access to a stranger that can potentially provide needed expertise or advice, there has been little theoretical work on how individuals actually go about this process. I refer to the process of gathering complex, changing and potentially equivocal information, and comprehending it by connecting nuggets of information from many sources to answer vague, non-procedural questions as the process of ?sensemaking?. Through a study of 81 fulltime IBM employees in 21 countries, I look at how existing models and theories of sensemaking and information search may be inadequate to describe the ?people sensemaking? process individuals go through when considering contacting strangers for expertise. Using signaling theory as an interpretive framework, I describe how certain ?signals? in various social software are hard to fake, and are thus more reliable indicators of expertise, approachability, and responsiveness. This research has the potential to inform models of sensemaking and information search when the search is for people, as opposed to documents.