How many tests does your software need? One test per single requirement/user story, or more? Testing everything is unfeasible (unless you have 42^10 good testers and 10^42 years at your disposal): you definitely need a criteria to identify what shall be tested and how deeply. Risk-based testing helps you to find the most important and risky parts of your software and to focus your attention and limited time on them. In this talk Daniele presents risk-based testing that leverages some powerful and very popular tools that a tester toolbox shouldn't ignore.