In my thesis I examine the possibility of increasing the efficiency of databases through normalization processes. To this end, I review the theoretical background of normalization procedures. I examine the errors, redundancies and possible types of anomalies that can occur in data models and in database operations. The system in need of improvement is a MySQL based database made and used by a computers and laptops servicing and trading company. In this database they keep records about laptop displays and laptop models, and by compatibility lists it can be determined which types of displays work with whitch types of laptop models. There has been a decision that a webshop is to be made in order to sell laptop displays based on this database. The original ad-hoc made database did not prove to be a reliable background for this kind of system to be operational. To repair the database, first I had to comprehend its structure, and the role of stored data, their relations with each other. Then I perform normalization steps trying to create an improved data structure appropriate to the customer's wishes. I discuss the existing data lossless information transfer of to the new structure. After that I test the modified database, and I carry out test measurements. The database operations and measurements have been done using phpMyAdmin tool. By the end of my thesis I would like to show by the results of the measurements, that by performing normalization a more functional data structure has been created, because although the alterations has measurable negative consequences, the positive effects significantly outweigh them.