BACKGROUND Previous reports have shown significant delays in treatment of in-hospital stroke (IHS). We developed and implemented our IHS alert protocol in April 2014. We aimed to determine the influence of implementation of our IHS alert protocol. METHODS Our implementation processes comprise the following four main steps: IHS protocol development, workshops for hospital staff to learn about the protocol, preparation of standardized IHS treatment kits, and obtaining feedback in a monthly hospital staff conference. We retrospectively compared protocol metrics and clinical outcomes of patients with IHS treated with intravenous thrombolysis and/or endovascular therapy between before (January 2008-March 2014) and after implementation (April 2014-December 2016). RESULTS Fifty-five patients were included (pre, 25; post, 30). After the implementation, significant reductions occurred in the median time from stroke recognition to evaluation by a neurologist (30 vs. 13.5min, p<0.01) and to first neuroimaging (50 vs. 26.5min, p<0.01) and in the median time from first neuroimaging to intravenous thrombolysis (45 vs. 16min, p=0.02). The median time from first neuroimaging to endovascular therapy had a tendency to decrease (75 vs. 53min, p=0.08). There were no differences in the favorable outcomes (modified Rankin scale score of 0-2) at discharge or the incidence of symptomatic intracranial hemorrhage between the two periods. CONCLUSION Our IHS alert protocol implementation saved time in treating patients with IHS without compromising safety.