Modeling is a statistical technique whereby you determine which pieces of data in your customer or consumer database explains the behavior of your consumers and costumers. People who work in data modeling are called data modelers and they work together with businessmen, as with possible users of the informational system. Data modeling would be a process of getting to know the data you are using, and the final product is the data model. There are several types of data modeling and different phases. Data requirements are seen as conceptual model and before being modeled, they are discussed with business stakeholders. When everything is settled the model is transfered to a logical data model. It collects the way a data should look like in a database. Afterwards the logical data is being transfered into a physical data model. There are various reasons why data modeling is highly recommended. To help IT engineers, managers, programmers, testers, manual writers etc. To easily manage data as a source. To design databases and database warehouses. Data modeling can be done at any time, and in any phase of a project. It should be seen as a live document which can be changed, edited, expanded, reduced at any time. It is not a finished product in this time of constant changes in technology.