The objective of this study was to estimate 1-, 2-, 5-, and 10-year conditional survival (i.e., the net probability of surviving to these dates given that the patient was still alive 0, 1, 4, and 9 years after diagnosis, respectively) for the most common tumors (15 in women and 17 in men). The annual net probabilities of death based on the time elapsed since diagnosis were also estimated. Data collected by the Francim network registries for the period 1989–2013 were used for this study. A flexible modeling strategy for excess mortality rates was employed. This strategy involved selecting a model using the Akaike criterion from a set of "candidate" models that differed in terms of the type of effect of the year of diagnosis as well as its interaction with time or age. The analysis was performed separately for each site in men and women. All estimates were provided by age. This study shows that the values of the 5- or 10-year cumulative net survival (the most widely used and reported indicator) result primarily from critical periods during which the probability of death is very high, and that outside these periods, the probability of death returns to lower values. These critical periods generally occur at the onset of the disease; however, for half of the cancers studied in men, a patient still alive four years after diagnosis still has a greater than 10% probability of dying within the next year. For women, particularly younger women, the situation is different and apparently more favorable. The risk of dying during the fifth year exceeds 10% for only a small fraction of the cancers studied. The highest risk often occurs immediately after diagnosis. However, for certain cancers (ENT, melanoma, cervical, ovarian, breast, central nervous system), the highest risk is observed around the first or even the second year, particularly among young people. Conditional net survival is an indicator reflecting the dynamics of excess mortality rates, complementing conventional net survival, which is a cumulative indicator. We can expect to see changes in conditional survival due to earlier diagnosis as well as the development of new treatments.