Background The survival of in-hospital cardiac arrest (8–25%) has not changed substantially in the past. Until now, most hospitals in Germany had no standardized protocols available for a course of action in case of emergency, and there are no continuous registry data for in-hospital cardiac arrest and survival. Objective Our aim was to improve survival and receive outcomes data, so we implemented a structured hospital-wide automated first-responder system in the hospital. Here our 5-year experience with 443 emergency calls is outlined. Methods Throughout the hospital, 15 automated external defibrillator (AED) “access spots,” which can be easily reached within 30 s, were identified. AEDs were then installed at these locations (Lifepak 500 and Lifepak 1000, Medtronic equipped with a Biolog 3000i portable ECG monitor). At the same time, a training program was initiated in which the employees of the hospital participated once a year. Participants learned how to apply and activate an AED in case of cardiac arrest even before the designated Cardiac Arrest Team arrived at the scene. Results A witnessed cardiac arrest event was confirmed in 126 cases. In 56 of the 126 cases, the primary arrest rhythm was either ventricular tachycardia or ventricular fibrillation and the AED delivered a shock. In this group, spontaneous circulation was reached in 44 cases (79%) and 23 patients (41%) were discharged. In 44% (24 from 55 patients) of the cases, a shock was recommended by AED and delivered by the first responders before the rescue team arrived. Conclusions The first-responder AED program successfully gave training lessons to the hospital staff. The training included how to initiate the cardiac arrest call, how to use the AED, and how to start immediate resuscitation. As a result, a higher survival rate after in-hospital cardiac arrest can be accomplished.