LAUSR

While current meeting tools are able to capture key analytics from both text and voice (e.g., meeting summarization), they do not often capture important types of conversations (e.g., a heated discussion resulting in a conflict being resolved). We developed a framework that not only analyzes text and voice, but also quantifies fundamental types of conversations. Upon analyzing 72 hours of conversations from 85 real-world virtual meetings together with their 256 self-reported meeting success scores, we found that our quantification of types of conversations (e.g., social support, conflict resolution) was more predictive of meeting success than traditional voice and text analytics. These new techniques will be essential to uncover patterns in online meetings that might otherwise go unnoticed.