We propose a new view on yield stress materials. Dense suspensions and many other materials have a yield stress-they flow only if a large enough shear stress is exerted on them. There has been an ongoing debate in the literature on whether true yield stress fluids exist, and even whether the concept is useful. This is mainly due to the experimental difficulties in determining the yield stress. We show that most if not all of these difficulties disappear when a clear distinction is made between two types of yield stress fluids: thixotropic and simple ones. For the former, adequate experimental protocols need to be employed that take into account the time evolution of these materials: ageing and shear rejuvenation. This solves the problem of experimental determination of the yield stress. Also, we show that true yield stress materials indeed exist, and in addition, we account for shear banding that is generically observed in yield stress fluids.