Training is an attractive solution in times where economy experiences a decline in the supply of skilled workers, which is the current trend in the Canadian construction industry. Regardless of its form, employee's training is expensive, so its benefits should be carefully assessed to assure a proper return from the investment. Research offers several methods, which can be used to measure the outcomes of training. Neither of those methods includes the non-linear changes of performance due to the "learning on the job" effect, which can be depicted by an employee's progress curve. This paper explores various training assessment approaches and offers analytical decision model. The model can be used to evaluate the impact of employee's progress curve on the time after which the benefits of training balance its costs. The model is illustrated with a real case of a construction project, in which cost effectiveness of hiring a senior PM is compared to training of a senior and a junior PMs, who are with the company for a number of years. The results of the study demonstrate that training of a senior PM is the most beneficial option.