When you use the Event Sourcing pattern in a .NET application, your data source just consists of persisted events. You don’t likely have a classic relational data store; all you store are events, and you store them sequentially as they occur in the domain. As you can guess, persisting events instead of a domain model has a significant impact on the way you organize the back end of the system. In this talk, we just develop a mini-ERP application that works out of distinct command and query stacks and persists just events. We also discuss how to rebuild state from events and see how to manage snapshot in order to speed up OLTP performances. Overall, this session presents a concrete example of an application architecture that for its inherent simplicity and maintainability is gaining momentum whether you have a complex business scenario to scale out or a just a CRUD system a bit more sophisticated than plain CRUD.