I introduce a new technique for directly designing control algorithms starting from input-output data collected in a "one-shot" experiment performed on the system to control and without requiring any identification step. I show that such design boils down to the solution of data-dependent semidefinite programs, which can be very efficiently solved