WiFi-based contactless sensing has attracted a tremendous amount of attention due to its pervasiveness, low-cost, and non-intrusiveness to users. Existing systems mainly leverage channel state information (CSI) for sensing. However, CSI can only be extracted from very few commodity WiFi devices through driver hacking, severely limiting the adoption of WiFi sensing in real life. We observe a new opportunity that a large range of new-generation WiFi cards can report another piece of information, i.e., beamforming feedback matrix (BFM). In this paper, we propose to leverage this new BFM information for WiFi sensing. Through establishing the relationship between BFM and CSI, we lay the theoretical foundations for BFM-based WiFi sensing for the first time. We show that through careful signal processing, BFM can be utilized for fine-grained sensing. We showcase the sensing capability of BFM using two representative sensing applications, i.e., respiration sensing and human trajectory tracking. Comprehensive experiments show that BFM-based WiFi sensing can achieve highly accurate sensing performance on a large range of newgeneration WiFi devices from various manufacturers, moving WiFi sensing one big step towards real-life adoption. AP. For example, when there are four antennas at the AP side, there are four BFM elements.