This paper examines a problem of worker misallocation into jobs. A theoretical model, allowing for heterogeneous workers and firms, shows that job search frictions generate mismatch between employees and employers. In the empirical analysis, the British Household Panel Survey (BHPS), the UK household Longitudinal Study (UKHLS) and British Cohort Study 1970 (BCS70) data are used to measure the incidence of mismatch, how it changes over time and whether it can be explained by unobserved ability. Results show that (i) the incidence of mismatch increases after the Great Recession. (ii) Individual transitions to/from matching take place due to workers' occupational mobility and over-time skills development. (iii) Employees can find better jobs or their mobility occurs earlier than the aggregate change of skills. (iv) Controlling for individual heterogeneity, measured by cognitive and non-cognitive skill test scores throughout childhood, does not decrease the incidence of mismatch. This suggests that unobserved productivity does not generate mismatch in the labour market.