This paper deals with empirical matching functions. The paper is innovative in several ways. First, unlike in most of the existing literature, matching functions are estimated not only on aggregate, but also on disaggregate levels which is unusual due to the scarcity of appropriate data. Moreover, the unique data set used allows to distinguish inflows into jobs by sources. Results for different measures of flows found in the literature can therefore be replicated using a single data set. Labor markets can be disaggregated by occupations, rather than by industries or regions. Furthermore, disaggregation is possible for age and educational groups. The paper allows detailed insights into the pattern of frictions in labor markets, on mismatch and labor market tightness, and therefore provides valuable information necessary for the conduct of labor market policies.