Smart Eye is a research project focused on developing and operating, in real-time, an innovative system, that allows the visualization of invisible, known, covered by the ground or by constructions, monuments and finds in an archaeological site, using an Augmented Reality (AR) environment. The visitors of an archaeological site will be able to observe the covered antiquities (3D models and descriptive information) on the screen of their mobile device (smart phone and/or tablet). The system integrates mobile positioning and pose estimation technologies, AR algorithms, Geographic Information Systems (GIS) and spatial databases. The AR application is implemented by IT scientists and is subsequently evaluated in the field to identify both technical problems (e.g. are files uploaded at high speed?) and problems identified by the average user of the AR application (e.g. would it be better if the button to display the 3D model of the archaeological excavation was larger on the screen or of a different color?). The paper will present, on one hand, the validation and evaluation protocols according to the relevant literature, and on the other, the exact evaluation methodology of the Smart Eye system. In addition, the problems identified and the way they were solved during the 1st evaluation of the application will be presented, as well as the problems identified during the 2nd evaluation.