We hear quite often the statement that behind "everything" there is a database. At first it may seem an exaggeration, and in a way it is, but it is true that almost every electronic tool we use contains or connects to a database, physical or in the cloud, more or less complex. Databases surround us and store each and every piece of data continuously. Our cell phones collect the steps we take, who we call, where we connect. A simple search in our browser will show us the vast amount of information that websites collect about us, our habits and habits. This stored data has become the basis of artificial intelligence (AI) and what has allowed companies like Google or Amazon to become the giants they are today. They have made the saying knowledge is power a reality. In the world that concerns us here and now, clinical and educational research, we have to pay special attention to the collection of data. They are of a sensitive nature and may involve a violation of the rights of study participants. For obvious reasons, the harmonization and anonymization of the data we collect must be an indispensable part of our work, not only for ethical reasons, but also because we are bound by the LOPD-GDD. We will focus on the questions we should ask ourselves before thinking about data collection: a) a correct database architecture, b) an adequate data dictionary, c) the balance between the technical ideal and the user ideal, d) which database to use and e) harmonization and anonymization of databases.