Shimer (2005a) claims that the Mortensen-Pissarides search model of unemployment lacks an ampiflication mechanism because it cannot generate the observed business cycle fluctuations in unemployment given labor productivity shocks of plausible magnitude. This paper argues that part of the problem lies with the correct identification of productivity shocks. Because of the endogeneity of measured labor productivity, filtering out the trend component as in Shimer (2005a) may not correctly identify the shocks driving unemployment. Using a New-Keynesian framework with search unemployment, this paper estimates that close to 50% of the Shimer puzzle is due to the misidentification of productivity shocks. In addition, I show that extending the search model with an aggregate demand side remarkably improves the ability of the standard search model to match the moments of key labor market variables.