By designing tests that address only the highest risk areas.

By identifying known risks in defect reports.

By prioritizing tests according to which product risks they address.

By calculating the likelihood and impact of a risk, the probability of finding defects is increased.