News sentiment has been a hot topic in the quantitative investment space for a number of years now, and in our past research we have studied how signals derived from this unique data source can add alpha to quantitative investment strategies. In this research, we turn our attention to a new frontier: web sentiment. Using a new database of machine-readable web sentiment data, we compare and contrast the efficacy of signals constructed using news sentiment and web sentiment. A particular area of interest to us is the difference in the aggregate sentiment derived from these two content sources – news sentiment is derived mainly from content targeted at professional investors, whereas web sentiment casts the net wider. Our hypothesis is that there will be differences in the sentiment derived from these two sources, and that there may be opportunities to exploit these differences in systematic investment strategies.