The stress-based approach with experimental S/N curves (stress amplitude vs. number of cycles) is a well-known approach for lifetime prediction of components, also in the composite industry. The problem with this so called "Wöhler" approach is that only final fatal failure of specimen can be described. But for a real definition of end-of-lifetime the identification of failure mechanisms and the corresponding sequence of effects are essential. Besides that for embedded layers, the initiation of first cracks does not lead to failure of the whole specimen. Within this work two different layups with carbon fibre reinforced polymers in UD 90° respectively embedded UD 90° within UD 0° direction was used. Starting from these two layups, a simulation was done for different end-tab-materials and several taper angles to get an overview on the lowest risk factor for breakage at the tab/specimen transition region of the specimen. Regarding the simulation results, specimens were produced and different end tabs were added. Quasi-static as well as fatigue tests were performed and compared based on the calculation with classical laminate theory. Also a comparison between simulation and testing of the crack region for the UD 90° specimen was done. With this work, a base for further variation of specimen geometry was established to monitor the initiation and propagation of (micro-)cracks in the embedded UD 90° layers with different techniques.